@Shades: Ah, thanks for that input, though I suspect that you might have been rather jumping to conclusions.
As is my wont, I've been quite thorough in checking the consistent repeatability of the things I reported. I assure you that I was not intending "...holding ProcessTamer responsible for this kind of misbehaving software", since none of the software is "misbehaving" - or, at least, not as far as I am aware.
It's very curious. I wondered whether the newer OS(es) had inadvertently caused some processes to be somehow partially transparent to PT - not totally transparent. It's as though PT can't get a handle on some processes. They are sort of "untouchables".
In testing (which I still have running), I have found that PT correctly kills those other Google processes you mention, and shuts them down again if/when they are auto-restarted a short while later. No problem. PT displays an alert each time it shuts down or modifies a process. Tedious, but at least it tells me what PT is doing, and when.
The first two examples I have given are simply that - two examples of a failure to "kill" by PT. PT does not display an alert that either process has been killed. Neither of them flicker out and then return in ProcessHacker, and the PID remains the same, so it's definitely not the case that they are being killed and then restarted without my realising it.
In the case of SeaPort.exe, for years PT used to successfully kill SeaPort.exe, and it would stay shut down (not restart), but no more.
The third example - the failure to modify the xplorer² process - is odd, as PT displays the process in its list of processes, showing its Priority as "Normal", and (amusingly) next to it the Explicit Rule "Force High". PT does not display an alert that the xplorer² process priority has been changed. I have tried, and PT can't seem to kill that process either.
By trial and error, I have discovered a couple of other "untouchables" - processes that I would normally not wish to mess with, but which, when I tried, PT seems unable to change or kill - and it so far seems to be a consistent and repeatable rule that if PT can't kill a process, then neither can it change that process' priority, and vice versa.
Maybe there is some simple explanation that I am overlooking here, but, as I say, it's very curious.