Basically what I think you're saying is...
Fly the SCA back towards your track from the 'new' drift-corrected heading rather than the 'old' one that got you off track in the first place?
Yes.
True that if the errors are small initially then the difference will also be small and that many errors in the system also come into play.
However, in order to be a valid method of "regaining planned track" then it must at least in theory acheive the aim. This is impossible unless the initial error is corrected (the planned track is paralleled) first.
Here are some examples that while extreme, demonstrate the point;
The first example is where the error is 40 degrees - the resultant of applying SCA and then track error correction is that the aircraft parallels the planned track throughout.
The second example (even more extreme) shows that the SCA track takes the aircraft further away from the planned track.
However:
The above example shows that even taken to the absurd (but not unheard of!!) error of 120 degrees, correcting the track error and then applying the SCA will
always bring the aircraft back onto the planned track.
So fix the error and then SCA back to planned track is a valid use of SCA to regain planned track while apply SCA to the track being flown in error followed by correcting the track error will never do better than a small error and could make the error worse.
I am not suggesting a major change - just a simple change to how SCA is being used which makes it 100% work. The only thing that needs to be changed is the sequence of actions - which make it foolproof (in theory)!