I think this is fair and is what data shows; regional services already mostly recovered, longer distance/inter-city is rising reasonably fast (it's imbalanced by route and being summer, W.Country services the busiest) and local London services (Hayes and such) still remain fairly busy around peak/pre-peak times. Its the commuters from Reading / Didcot / Newbury / Oxford who are staying away.
What the GWR data shows, as was explained this morning by GWR's interim development director during the Cotswold Line Pronotion's Group online AGM is that overall traffic is currently running at around 20 per cent of the normal volumes (same goes for XC and WMR, the meeting also heard). What happens in September is going to depend heavily on whether the return to school does happen smoothly, which is clearly far from guaranteed.
It is indeed busy in the tourist areas of the South West right now - not exactly going to be a revelation in August unless we were still in full lockdown mode - but that is not the case elsewhere on GWR and what the long-term trends are going to look like is still anyone's guess.
But one thing I can predict is that it will still be the case that for large parts of the year there will be no justification whatever - based on the number of passengers travelling west of Plymouth - for all London-Penzance services to be formed of nine-car IETs past Plymouth.
Unless, of course, people think that getting rid of the recently-introduced 2/3tph frequency through Cornwall is a good idea, so lots of empty seats can be carted up and down on IETs as part of a reduced frequency timetable. As that is what would happen to balance the books on the costs of operations in Cornwall, as Clarence Yard has explained here over and over.