Well lessee, during my "active" period as a Crew Chief, my drivers won something like 41 races in 69 starts, 2 SARRC Championships and 4 ARRC's. That doesn't mean I know anything, but IMHO (and I mean that sincerely) here's my take on pyrometers/gages.
I use a pressure gage to set the first cold pressure of the day, and each time we mount another set or rotate. I record pressures after the tires have cooled (as well as the ambient temp) for setups that we are happy with. In 15 years of ITS I may have checked pressures hot a couple times just for yuks. They mean nothing to me. The pyrometer is my primary tuning tool. In fact, the last two years I worked regularly I never unpacked my camber gage, and couldn't tell you what camber we were running. In the first few years with the Z's, however, we tracked and recorded so much data I think we should have owned stock in Weyerhauser. After enough years on the same tracks you sorta learn what you don't need.
In the bias ply Hoosier days a 1/2 lb of air made a significant and repeatable difference in the temperature distribution. Ditto my days on slicks with sports racers and prod cars. As already noted, the bias of temperature for optimum grip varies with suspension and tire type. Pre IT radials I'd shoot for even or 10 degrees max depending on which car we're talking about. On radials its more like 25 degrees hot to the inside. We seem to get a lot of benefit from camber thrust in radials, and curiously that kind of bias still results in pretty even wear (although wear in general has a LOT to do with shocks). Radials are not near as sensitive to pressure as bias ply's, although being basically a race carcass the Hoosier is pretty close. You're not losing because your tire pressure is off a pound or two.
I never make a setup change until two sessions produce the same temps, unless of course we have something way wrong (this is usually indicated by the driver running over your feet on the way to the paddock while your in pit lane waiting for him to stop for temps). If you think temps only reflect the last couple of corners then you're not sticking the probe in deep enough. Because of the huge camber most IT cars run now (eh, the ones with not enough spring and shock anyway
) the corners of the tires usually have a pretty big temperature gradient so to be repeatable you have to be careful and consistent about which locations along the tread you're taking temps. I always measure hottest tire first, followed by next hottest etc. The hotter the tire, the more heat loss while you're measuring.
While I can't say enough good things about Bruce Foss and Rodney Perry of Hoosier, I always do my own temps even when I also have them do them. They're usually pretty close to right.