I think there were:
The original train that got snared near Acton, and another that left Reading a little after that and got to Stockley Junction and hung about for ages before going back. Both had 80xs auxiliary power from their own engines.
Two HEx 387s, operated by GWR▸ , which happened to be close to Acton when power was lost. That should have made it relatively easy to get to them quickly, perhaps using the same staff for both, and keeping in touch with those near the first train (and Andrew Haines?).
All the other trains were TfL» 's 345s, spread out over the route - five, as I recall. Getting to all of those was the big challenge. I'm not sure how much TfL's own staff were able to contribute to the process.
Of course if it wasn't a GWR strike day there would have been more than two of their services.
Once the location of the de-wirement was known the isolation could have been shorten back, trains could have then at east moved to the next station or set back to one.
The limiting factor preventing the Electrical Control Room from shortening back remotely is not all of the required remotely controlled earthing devices between Paddington and Maidenhead have been fully commissioned as yet, therefore the Earthing would need to be done manually by the
OLE▸ Maintainenance teams which takes time.
It is a requirement under an emergency isolation of the 25kV OLE the power is discharged between Neutral Sections, until an isolation and Earthing is done in the local area of the incident. Neutral Sections are located at Maidenhead and Royal Oak (Elizabeth Lines)