Hope you're not letting the AFR numbers scare you. But when you write 'right AFRs' it seems like you're chasing numbers. Think of AFR numbers not as goals, but reference points, clues that help you figure out what is going on. Every engine will have slightly different needs. Your goals should be most economy and most power.

Urich & Fisher's Holley Carburetors and Manifolds provides a small section on part throttle tuning. It's aimed at road racing but the test methodology is to hold a vacuum in 3 or 4rth gear and measure the time it takes to accelerate over a set distance. Then do it again at a lower vacuum (more throttle) and again lower yet. Take it down to the PV opening. Then do it all again with richer or leaner jets until you get the best times. If the best jets don't match the jetting for best WOT, then adjust the PVCR.

That's pretty tricky to do safely on public streets even with a data logger, at least around here. However a dyno that can apply an eddy current load can work toward a similar test program. I tried it on inertia only but I don't think its terribly valid and its hard to control the throttle because the load isn't higher enough. There aren't too many dyno operators that are going to be this sophisticated about dyno testing and know carbs.

Simplest approach is to get best mph on the 1/4 mile. The nice thing about AFR monitoring is you can see if its changing during the WOT top gear. If it is, look at the HSABs. Then see if you can go leaner for highway cruise with the primary main jets. It won't ping at light throttle, it will lose power or rev up and down. At that point, you got to slow the car down to 50 or 55 mph and when you get back, jet richer. If you're doing WOT on an inertia dyno, jet a step or two richer before going out to the real world. I didn't one time, and thats how I got to experience what too lean feels like.

- I'm not sure if it the table I posted was clear. Throttle position is roughly a diagonal line on that table. A throttle position sensor is certainly another good tool in the datalogger arsenal. I'd like to add one.