While I haven't asked and therefore cannot say for certain what is referred to in this paper, there are people working on the Mexican taxa (i.e. not just the previously mentioned 2004 publication). As far as I know, no-one has ever split the populations on Trinidad and Tobago into separate taxa, and no evidence seems to support such a move.
The results in this paper are interesting, but overall follow the expected picture. I do wonder what various authorities will do with argenticinctus, as several Neotropical authorities have argued for requirering biological species to be monophyletic (despite strictly speaking not being a requirement in BSC). I also have to wonder where exactly the shift between the lessonii and subrufescens groups occurs in Panama. Quite some time ago I stumpled upon this individual labelled as being conexus (i.e. part of the subrufescens group), but being puzzled about its lessonii-like plumage I removed that label, and this paper appears to support my suspicion.
http://www.birdforum.net/gallery/showphoto.php?photo=111178
Regardless, based on the map in the same paper, it really should be conexus (= subrufescens, per new taxonomy) in this part of Panama. Despite this being among the most commonly visited regions in Central America, xeno-canto currently doesn't have useful recordings that can help resolve the issue, and my own Panama experiance is very limited, but based on above photo it could well be that the two groups actually come into contact in Panama. Even if not in actual contact, they're separate by ~100 km
at most and with no clear ecological barrier between them, as conexus (= subrufescens) has been recorded at least as far west as the Canal Zone. While I have no doubt about their validity as separate species, it would still be interesting to see what happens if in contact.