Is there measurement that correlates with cohesive/pinpoint imaging?

I'm afraid all the components in the Snell A/III are outclasssed today, but for imaging and impact, amazing.  The SF Stradivari are much newer and deal with the diffraction issue very differently than a large box does.  Outstanding.

because a large baffle destroys time information while functioning as a mechanical averaging machine

Wait, what? Some of the best sounding, best imaging speakers I've ever heard are the Sonus Faber Stradivari and Snell A/IIIs. 

You can make a wide baffle with low diffraction, and if you can go for it. :)

Best,

E

Hey OP:

You really should at least start to read about HRTF. It will help you see the problem of imaging in a whole new light, and why it is so hard to measure.

  Do phase alignment/time alignment correlate?

Not in my experience, but clearly the fans of speaker brands like Thiel and Vandersteen feel strongly they offer special features you cannot achieve otherwise.  Personally I vote with the literature, that says we aren't that sensitive to phase or time alignment. We _are_ sensitive to badly matched drivers however.

There's also some research saying that making the frequency response dip around 2.4 kHz enhances the experience with the trad off being a neutral frequency response.

It is also important to include the room in these discussions. Not only does controlling reflections matter but enhancing diffusion in areas also helps a great deal.  I find that imaging is enhanced in the same plane as the room treatment.  Want wider?  Treat the sides.  Want taller?  Treat the ceiling.  Want depth? Treat behind the speaker and behind the listener.

The best speaker in the world can be ruined by a poor room.  Conversely, some speakers reduce the interaction with the room, like line sources, planars, horns and open baffle.

AFAIK, there is no 1 number for this, but if you want to jump into a fun rabbit hole, research "Head Related Transfer Functions" or HTRF for short. :)