Great to have you Bert!
I have a 332B calibrator which is in need of repair -- it works, but if you log the voltage you will see that it will stabilize for a few hours, and then start wandering around by a few hundred millivolts for a while, then stabilize again, etc. I'm not certain, but I believe the 3330B may be based around a similar circuit, but is a bit younger. It may suffer from the same sorts of issues (I.e. You are probably going to need to replace a few capacitors here and there).
Do you have data logging set up with any of your meters? I'd say your first step is to evaluate the stability of your 3330B by logging it for 24 hours or so.
You've got a great start here -- the 34401A is probably the best value available for budding volt nuts who haven't yet ascended to a higher plane of existence (by purchasing a 3458A). The 34401A gives 7.5-digit resolution over its data logging interface, supports 10GOhm on the 10V range, is the easiest to get started with data logging (use "talk only" mode on the serial port), and is both plentiful and old enough that it can be had for under $250 on ebay if you are patient and lucky.
Hook up your 3330B to one or both of the 34401A and 3457A (using both would help deduce if any strange behavior is due to the 3330B or due to one your meters), take some data, and then make a plot to see where you are at.