Well that is not what the manual says (Contingency Aborts 21007/31007) at page 25: for the 2EOUT C.A. case they talk about FAST ET sep which occurs 5 seconds after MECO is commanded (or 1 sec after SSME shutdown) so that means no mated coast at all.
Maybe FAST Sep was not always the procedure to follow and in some case they would perfom a manual SEP at a later stage..
Looking into the aborts workbook, you are largely correct.
In practical terms (flying a simulation), I believe you're still fine.
What the various ET-sep techniques are supposed to solve is a re-contact of the ET with the TPS by making sure the airstream at the separation point helps you get away from the ET rather than sucking you towards it.
Unless you have a
very high-fidelity simulation of the aerodynamics of separating Shuttle and ET
and your simulation includes TPS damage by touching the ET again, the issue is a bit academic - in a 2EO scenario you do have a buffer of a few ten seconds till aerodynamic forces get really strong and the ET becomes unmanageable trouble (assuming your simulation actually includes the aerodynamics of Shuttle mated with the ET vs. just Shuttle).
So while this presumably is pretty important in practice, I''m going to lean out of the window and say that to the degree we can simulate the ET-sep inside the atmosphere a few seconds earlier or later won't matter a thing.