Someone posted awhile back that there was something done to affect "FM Filtering." I have no clue if it was in the front end or intermediate stages. At any rate, I suspect it is FCC scanner approval that would be the biggest roadblock to substantial hardware changes.
Any changes were minor. The main upgrade was a CPU with more memory (so I assume -- can't get anyone to open the shields for me). If anyone wants to look for me, this will be the big TI integrated circuit.
But I really haven't experienced what you are describing. I mostly use a 1080 in the truck and I receive tons of stuff. I can reliably get C4FM Project 25 twenty to thirty miles from the transmitter. And that's with the radio sitting on the console directly above several of the vehicle's computer systems, including the radio. I use just a $10 800mHz antenna. Nothing external (new truck is aluminum); nothing fancy.
Of course, reception of marginal systems for the location or of simulcast can vary by location. I can't reliably receive some systems that I can at home; on the other hand, in certain spots I pickup systems that don't work well in the apartment.
Another consideration is programming. Are you scanning too many objects? If your scan takes over a minute to complete, that's going to cut down on what you hear (on the basis that much of that scan is looking for frequencies out of range). It's certainly going to cut down whether you hear the more interesting traffic. How do you have your scanners programmed for the road tests?
Finally, what types of systems aren't working? Is this really about simulcast problems?
If you take apart the scanners, you'll find that all of the modern ones are all very similar. That's why this reception battle is really interesting to me. I think it's more related to execution (i.e. good shielding) and programming.
The main differences between the major models (i.e. Whistler vs. 536 vs. HP vs. 996 and variants) are the processors, software, user interface, display, and appearance. The software and user experience is probably the most substantial difference between any model line and certainly what most people notice and complain about. The rest is pretty much the same. Even UPMan said awhile back that ~90% of the components are the same (referring to his company repairing scanners of other makes).
How many American engineers are actually working on scanners? Is any of the engineering done in our country?
I'm not sure there is such thing as "too sensitive." Perhaps what you are really looking for is higher specificity (i.e. rejecting unwanted emissions). Many folks on here wish their Whistler had better sensitivity, and their radios of other makes receive traffic that Uniden and Whistler fail to bring in. Certainly, though, Whistler and Uniden are at a severe disadvantage due to federal regulations. And I'm not sure if anyone other than UPMan has actually taken any action to eliminate the asinine regulations.