Jeff Evans
I like the second half a little better. I also hear the change as slightly broad in nature. To me the sound is a little less cluggy in the low mids and the upper mids/highs are slightly pushed forward. Sounds like a tiny bit less of the notes and a touch more of the distortion etc..
Thanks for listening!!! What you describe hearing are the precise results this effect is supposed to produce.
But maybe you don't need a complex comb filter to achieve a similar result. I feel I might be able to get it just the same with something like the Pultec EQ.
Well, you know how much I like Pultecs

but I don't think that will do the job here. Although the change
sounds broad in nature, the four notches are very sharp - IIRC the Q is around 20-24. This effect is somewhat guitar-centric so let me fill in the details, and while I'm at it, explain why arranging is not well-suited to fix what this technique fixes.
The open strings on guitars produce more level and sustain longer than fretted notes. I don't think it's a coincidence that the most popular chords for pop guitar (E, A, D, G) include at least one open string for the root and at least one additional fretted note for the root. As not all chords have voicings that encompass all six strings, the root is often represented by more notes than any other chord note. Interestingly a standard four-note D major chord (i.e., the root, not the 5th, is the lowest note) has only one open string and two root notes total, but that's perhaps why dropping the low E to open D is one of the most popular alternate tunings, if not the most popular.
Barre chords are also popular; although they don't have open strings, a typical major chord in the most common barre fingering has three fretted root notes - half of the total voicing. Furthermore, even solid body guitars resonate. When you have multiple root notes, not only do they tend to reinforce each other, but will excite open strings in that key if they aren't muted. (And of course, pianos are off the hook in terms of resonances.)
This is why arranging isn't the issue. There are only so many ways you can voice chords on a guitar, and music notation doesn't have instructions like "play the D a few dB more quietly if the voicing uses an open string and a piano is playing at the same time." So, the sharp filter prevents interference with nearby notes that tend not to resonate and aren't being played on open strings, which allows "taming" open strings and resonances if they become excessive.
If you line up the second half of the audio example and throw it out of phase with the first, you might be surprised at how little difference there is given the results. But what's left has a sharp, whistling sound, not the broader bump you'd hear with a standard EQ.
Thanks again for the comments, it's helpful to have an objective set of ears confirm what I'm hearing. I only post about experiments with results that I find useful (no need to embarrass myself posting info on experiments that fail), but my hope is that others will find them useful as well.