We seem to be having the same argument on two different threads (one about London - Manchester extensions and one about London - Leeds extensions), so I'm not sure which to post this in, but my basic thoughts are as follows:
This will always be a subject for debate. Obviously nobody *enjoys* changing trains, but then where do you draw the line to minimise the number of people who have to do so? A lot of this debate is focussed on Somewhere - Leeds - Kings Cross, but not all demand is *to* Kings Cross - we aren't debating direct trains from Leeds to Liverpool Street (even though it'd be handier for financial types heading to the City Of London and I'm sure that they don't like changing trains to get to the Square Mile just like Bradfordians don't like changing trains to get to Kings Cross
BR were quite canny at squeezing extra out of stock, e.g. there were Intercity diagrams from Edinburgh to Birmingham/ London that didn't leave Craigentinny depot until after the morning rush hour - this allowed BR to put an HST on a service from Aberdeen/ Inverness to Edinburgh, providing a lot more capacity than the "standard" service (and presumably allowing the train that would have run that Aberdeen/ Inverness path into Edinburgh to be used elsewhere to provide additional capacity on another line - e.g. HST replaces four coach train on the Aberdeen - Edinburgh service allowing the four coach train to replace a two coach train on a commuter service into Edinburgh from somewhere like Dunfermline). That's why the Highland Chieftan has the paths it does - it was originally the last HST out of Craigentinny - i.e. before the first northbound services arrived into Edinburgh from London to start providing subsequent southbound services - and the first HST back into Craigentinny - i.e. not a lot of demand from Edinburgh to London around 18:00 so why not use that stock to provide additional capacity at Inverness, even if there's precious little through traffic)
Given that LNER's demand flows are still fairly similar (in terms of "southbound in the morning and northbound in the evening" - peak extras etc), there will be some scope to run a commuter service into somewhere like Leeds with an 800/801 since it'd be better overall to fill a five/nine coach train with commuters from Skipton/ Harrogate into Leeds than run that unit as a fairly quiet London to Leeds service that leaves Kings Cross before 07:00. So I can see the attraction in using stock this way - there doesn't have to be a huge demand to get from Keighley to London (after all, what has London got that Keighley doesn't?) but it's more about "using a long train for Keighley to Leeds at rush hour is The Greater Good rather than keeping it in the depot until after the rush hour or beefing up the London - Leeds service in the morning peak. It doesn't even have to be a through service for London - e.g. you could use the stock for a Hull - Leeds service in the morning rush hour even though you're not going to run a Hull - Leeds - London service
One problem with that is that, whilst there's not huge numbers of Londoners trying to get into Leeds for 08:30, there will be a number of people coming from places like Wakefield into Leeds. So if you thin out the northbound service into Leeds in the rush hour (to provide additional capacity from Harrogate etc) then you have to accept that Wakefield passengers cramming onto a shorter Northern service is the price that you are prepared to pay. There's already a problem with Darlington/ Durham (where the peak service into Newcastle sees people crammed onto a short Northern DMU because there aren't many London/ Birmingham services that arrive into Newcastle until later in the morning
However, there are two other kinds of things being talked about on this thread (and the Manchester one).
Firstly, the "05:30" kind of token services that only exist to "put the town on the map" and give local politicians some kind of civic pride. Waste of time, disproportionate waste of resources too (e.g. it might only be one train a day that means stretching a diagram at the extreme margins but you're going to need to ensure enough staff are trained up on it - you're going to a lot of effort for minimal quantifiable benefit
Secondly, the "off peak" extensions. Tricky one. Sometimes you have a regular off peak service that a LDHS (Long Distance High Speed) train can work on - e.g. a Northern semi fast from Leeds to Harrogate that stops at the same stops that LNER use and takes an additional two hours for the round trip works out quite neatly for an extension of a London service. However, running an extended London service to Bradford Interchange or Skipton runs into the problem or whether the LDHS service on a busy line is meant to replace a Northern diagram and stop at Frizinghall/ Cononley or whether these stations miss out every couple of hours, since I doubt there's much space for the LNER service to be in addition to the existing Northern services through Shipley
To take a different Yorkshire example, there's two EMR services per hour from London to Sheffield and two "semi fast" services per hour from Sheffield to Leeds via Barnsley. Barnsley is one of the bigger places to lack a direct London service (it had one briefly under Midland Mainline). However, if you extended the EMR services to Leeds in place of the Northern services then you'd have to do something about Meadowhall (the platforms are too short for a 222 or an 810, so a "Barnsley - London" service would mean a reduction in services to Meadowhall)
And, if a five coach IET costs many millions of pounds then you've got to consider whether it'd be great use to always have one beyond Leeds (whether to Harrogate or Bradford or whenever)
Plus the complication of joining/splitting (e.g. if your plan is a ten coach train from London to Big Provincial City which then sees the front five coaches continuing to Smaller Place and then joining onto the next London bound train at Big Provincial City an hour or two later)
tl;dr version: It makes sense if you're using the train in spare time, but not so early/later that it's just a box ticking exercise - good connections at key interchanges (Leeds, Doncaster, Peterborough etc) are more important