I'd be really curious to see how your L+R speaker pair measurements and the overall measurement looks compared to your individual speaker responses, if you happen to get the time to take them.
I was kind of surprised to see
a) how different the dB increase was when playing L+R midbass vs. L+R dash speakers (my midbass speaker pairs gained about 6dB compared to individual speaker dB levels, but my mid/high speaker pairs only gained about 3dB when L+R speaker pairs are played)
and
b) how certain areas of the L+R responses didn't match the shape of the individual speaker responses - I had quite a few phase- or reflection-induced dips when playing L+R speaker pairs - which I then had to address as another whole round of EQ corrections. Then I had to do yet another round of EQ when all speakers were playing. The end result is that my individual speaker responses no longer matched the curve by the time I was done (but the overall response did). :-)
A few other things I recently noticed that surprised me a little:
1. When adding my rear speakers to the overall measurement, it didn't just make the overall response louder in the freq range that the rear speakers play - it actually changed the *shape* of the response quite a bit.
2. How much the overall response changed when I changed the delay time for the rear speakers. Using a delay of 10ms looked very different than when using a delay of 20ms, for example.
3. The change in response if I had my windows cracked just 1/2"! It caused areas of the bass response to lose a few dBs!
4. How different the response looked with relatively minor temperature/humidity changes - mainly with frequencies above ~2khz.
I'll be posting step-by-step measurements as I go through the entire process shortly.
I've never used separate target curves for each individual speaker before - I just let REW figure out the individual speaker curves from the overall target curve since it basically just "creates" the individual speaker curves from the overall curve automatically "on the fly" (depending on what you input into the "Target Settings" fields). Just different ways to accomplish the same thing I guess (as opposed to creating separate individual-speaker curves with Jazzis spreadsheet)?
I find it very interesting to hear how other people do things. :-)