Let me sum this up:
...
You then use process control strips from one specific vendor to fine tune your processing setup, which includes changes to chemical composition of the processing bathes and optional changes in timing and temperature.
Here's what I'd say: don't change the chemical composition, but CHECK whether you did something wrong, or if the chemicals are somehow defective. Since you probably don't have a way to check the chemical makeup, a possible test is to buy fresh chemicals of a reputable brand and see if they give the same result as what you first used.
Don't change the time - this is the least likely to be significantly in error. But DO confirm that the time is correct.
Try modifying the amount of agitation in your process, as this is the one parameter that is not easily quantifiable.
If moderate changes in agitation fail to get your control strip plots into spec, then consider the possibility that your temperature readings are in error. And do whatever tests might be feasible to check into this.
Remember that these processes were not originally designed for hobbyists, and are not necessarily easy to troubleshoot. It's a complicated system, and you might even say that it's surprising that home users are even able to get acceptable results. There's something of an art to effective troubleshooting, and some of the most efficient techniques are based on already having an in-spec process nearby, so you can do substitution tests.
Ps, now that I already advised NOT changing "the chemical composition of the processing bathes," I want to modify that. If you are running a replenished process there is the possible situation that you have some heavily exposed film which "exhausts" your developer faster than average. This actually HAS changed the composition of the developer away from spec, and the normal correction for this is to increase the replenishment rate. This of course is not a perfect solution, but it is usually a reasonable compromise, so I'd say it is fine to modify the chemical composition in this manner.
- Different densitometers give different readings with the same process control strip, too
This is not something that generally happens to any significant degree, provided that you are using the appropriate high quality densitometer. For color neg, I think the spec is for a "status M" response, which is well specified spectrally. A densitometer typically reads to the 0.01 place, with an accuracy spec'd to perhaps 0.02 (you should verify this, as it's off the top of my head. But if you read a control strip with a different densitometer, it's unlikely that they will differ by more than 0.03 or so (on some mid-scale patch). But even this is not much of an issue with control strips. If you read the procedures (see process control or troubleshooting in Kodak's Z manuals), you see that a preprocessed reference strip is included with control strips. So you are not relying on a densitometer for the correct absolute density value, you are using it to measure how far your test strp deviates from the "ideal" test strip.
Typical test strips are not necessarily made by the maker of the process chems,
You should ideally be using process control strips supplied by your film manufacturer, and in fact, you should be using the chemicals specified by that manufacturer. If a different chemical supplier claims compatibilty, the burden should be on them to show that this is the case. But again, you have to process to their specs - if you have the wrong temperature or time, well they are not to blame.