I think metro services might be on the way up though, to replace commuter trains. Crossrail was a huge success, and Manchester is showing aspirations for a Merseyrail style network. Merseyrail itself is also a great example of seeing metro service impact over the majority of those 50 years.
I can see other cities like Sheffield and Leeds looking to catch up with the other major cities.
Indeed, very local trains (London Overground) are the sort of travel which is likely to be useful.
The question comes, do we try and add extra stops to existing allow services to make this viable or do we run new services in the shadow of the existing services.
Any urban area with a population of over 1 million should be looked at first. However, over time smaller places could be looked at. Any where there's spare track capacity could be easy wins and could actually boost passenger numbers on the wider network by making rail travel easier to access.
We've already seen a trend towards this (for example the Exeter semi fast services which provide better connections to smaller places).
One of the examples I've given in the past for this is a Basingstoke to Ascot service (works require a grade separated junction east of Farnborough and South of Frimley to allow the service to run, otherwise would be only using existing track).
The slow lines between Basingstoke and Farnborough see 4 passenger trains (one only at Farnborough) an hour and maybe 2 freight trains, so there should be plenty of space to add in an extra service or two an hour, whilst the Frimley to Ascot line sees 2 passenger trains an hour.
Basingstoke is the limiting factor, and this is where further enhancements may be needed. Options include, Leisure Park gaining a siding with the metro services and the Basingstoke stoppers running to there to turn around not at the main Basingstoke station platforms, or electrification to Oakley, Overton, Andover or Salisbury to allow the Metro and Basingstoke stoppers to return around there.
Assuming that it only reached Basingstoke there's plenty of extra stations with decent (not big enough for a new line but a new station) population sizes and/or with places people wish to travel to which could be added. Basingstoke Leisure Park, Old Basing, Edenbrook/Elvetham (Fleet West), Southwood (Farnborough West, housing and business park), Frimley Green (at the new junction), Watchmore Park (Sainsbury's and a business park and not far from housing), Old Dean/Crawley Ridge, with others possible with future developments.
Whilst passenger numbers to/from each of those wouldn't be huge, and nor would the likely local ticket costs generated, if it generated no additional commuter or longer distance passengers (unlikely, as Southwood has one of BMW's UK major offices, which could see at least some passengers to/from the traditional rail network) each new station would need to average about 350 passengers a day (circa 250,000/year) paying £3 each to cover costs for a 1 every 2 hourly service (£200,000/coach lease costs, with lease costs being 1/3 of total costs).
That hasn't allowed for any increase in passengers from existing stations, for passengers going more than 2 stops, for the new journey options (i.e. Ascot to Basingstoke becomes a direct service, rather than a change, requiring a reasonable walk, at Reading as well as every station pairing between the two), none of these are going to be generating millions of passengers, however likely enough to justify at least an hourly service, given the existing stations a 3tph frequency.
With a 15ish/15ish/30 minutes service gaps is likely to see reasonable use. The preference would be 2 metro services to give a near 15/15/15/15 minute service gaps (i.e could be 12/18 in one half hour and then 18/12 in the other) however that would likely need wider benefits to be considered, especially given there would be need for a grade separated junction (one chord, Farnborough to Frimley, previously existed so shouldn't be too costly to build, it's getting back the other way which would likely need a new bridge under the Waterloo mainline and a chord over a lake being built which is where it could easily cost £150 million), however in an area where developments are likely to add to the already congested roads and rail use is fairy normal (there's a lot of collage aged rail travel in the local area, so making that easier from more places would likely cover a reasonable percentage of the uplift needed - say 20%)
Anyway it's those sorts of easy wins, where there's scope to add in services where there's local journey opportunities without impacting the existing network which are likely to see decent growth.
Even if a service doesn't serve the main city station but provides interconnectivity between lines which otherwise there's limited connections, it's likely to see fairly good use.