That's quite a well argued post. I think the situation is subtler than you suggest though.
I think there's a difference. In the places where trains are leaving passengers behind, there's basically no other choice for getting to the local large city (Manchester or wherever) - so people are going to have to go to the station in the hope of getting on a train even if they know it's difficult getting on. For the Chiltern line, that's not the case. Just about every station in the London area is within walking distance of a tube station that gives an alternative route into central London. So any suppressed demand won't manifest itself in big crowds for the small number of trains running - it'll manifest itself in people heading for the nearby tube stations if they want to get into central London - even if Chiltern would have been more convenient - or probably driving if their destination is not in central London but would've been served by Chiltern.
So the case for better metro services on the Chiltern line isn't about easing pressure on Chiltern itself: It's more about easing pressure on the tube lines (which may not be crowded at West Ruislip, but are overcrowded nearer central London - so any passengers you can get off them would be welcome), and also about providing more choice: Giving people the option to make journeys by rail that are currently only realistic by car or sometimes bus. For example, a link to the Overground at West Hampstead would allow a huge range of orbital journeys to/from NorthWest London. And we know from the success of the overground that an awful lot of people in London want to make orbital journeys that don't go into the centre.
Well yes, and that's going to be because, in general, long distance services are much more profitable than metro services. As far as I'm aware, metro services have huge benefits in terms of easing congestion in urban areas, but are rarely directly profitable. And I would think the reason runs like this: As an example, in half an hour, a 'mainline' service might travel about 30-40 miles. You can charge people quite a lot for taking them that distance in half an hour! But in the same time, a metro service might only travel about 10 miles. It'll cost about the same to run, because you're using a train and staff for the same amount of time, but you can't charge people very much for taking them only 10 miles in half an hour. Hence metro services don't make profits, and hence why a commercially astute private operator is not going to be very interested in running them if their franchise doesn't demand it. But that doesn't mean they are any less useful to the country and the economy as a whole! (That's all my own reasoning - I haven't seen it anywhere official, so I hope it's correct).
I think this is going to boil down to what each person thinks the railway should do.
The points you've made are valid if you believe the railway should leave no stone unturned and spare no expense to increase patronage even by a very small amount. I don't believe that. My starting point is the old American maxim: if it ain't broke, don't fix it.
As mentioned previously, there is ample evidence that commuters in North West London and urbanised South Buckinghamshire are content with their railway provision, which most certainly is not the case in many other parts of the country. Those commuters have not given their proxy to railway enthusiasts and have not requested anyone to be indignant on their behalf! The situation isn't broke and doesn't need fixing.
Your assessment of the economics of commuter railways is correct as far as it goes but does not include the significance of shared route costs or the critical effect of commuter numbers. Certainly in many cases, subject to shared route costs, the profit per passenger on a commuter railway is less than per passenger on long, Inter-city routes. However, if a TOC can make even a small profit per passenger on a commuter route, then, if the number of passengers is enormous, there will still be worthwhile profits.
Example 1: South West Trains out of Waterloo was until recently one of the two most profitable franchises, and commuter services provided a very large percentage of their income, possibly most. It is inconceivable that the longer routes to Portsmouth, Weymouth and Exeter produced such a surplus that it both financed a loss on commuter services and created a big overall profit. Instead, it is conceivable that the huge commuter market made a substantial contribution to that profit.
Example 2: C2C Railways between Fenchurch Street and Shoeburyness is almost entirely a commuter railway and has always been financially successful.
The lesson here to any TOC is that they should judge the scale of the commuter traffic they are likely to win. It is evident that Chiltern have made the calculation and have concluded their London commuter market is not worth the effort. As customer satisfaction with Chiltern Railways is high, the service is not broke and doesn't need fixing.
All of this is additional to the issue of Marylebone Station which has little spare capacity. My guess is that with lengthened platforms, Marylebone will be able to handle longer trains far more easily than extra trains. The cost of platform lengthening will be minimal compared with the cost of 4 tracking through London and across those Buckinghamshire hills.