The platform-test I made uses two scripts, but only one writes to the globals while the other reads them. The timing is more sensitive than having it all in one script, but if in order it goes MUCH faster. I agree with the update blocks never running simultaneously (which I've found to be the case, personally) - one would either stack to the next available time-slot or simply be dumped.
What I was wondering about this and what you said about the reg4Sing cancelling a previous though was does it cancel the onUp block while it's in progress or only if it hasn't run yet. EG: the 10sec delay in script 1 means it hasn't fired the onUp yet, thus gets cancelled. But if that 10sec delay expired and the onUp DID run, will it be nixed mid-stream? THIS is what I don't think is the case... but I haven't fooled around with calling 'cross-script' functions and variables (except globals, but that's not the same).