If I want to do that sort of processing to a small section of the track, I’ll often make a duplicate track, snip both tracks around the section,mute/delete parts so only the section to be processed plays in the new track and vice versa, then bounce and apply the processing to the new track thus avoiding doing any faffy automation and giving a lot more flexibility and ease of mixing also. Once convinced I’m happy I’ll bounce the 2 tracks back together if it’s a big project and I want to keep the track count sensible.
All this is much quicker to do than to type about! It’s basically doing just as you suggested - highlighting a small section by spinnining it off to a new track…