I can't provide any substantial advice on exact camera sensors. But I would like to offer my input on other mentioned aspects, as I feel we're about to slide into territory of personal preferences, but kinda packaged as if "universal" empirical fact. So perspective from the other side might be appropriate.
1: Not everyone *can* get comfortable with working through eyepieces. For some it's the viewport angle, and stereo/trinocular scopes with significant tilt-adjustable viewports are quite expensive. For some it's the limited work height/focal distance range at their preferred magnification (and resulting limiting height range between head and work surface). For some, including me, it's both. There's plenty more factors here too.
2: Image latency can be bad, over USB or Wifi. Horrible even. But over HDMI directly to a midrange screen, it's imperceptible for most people, single digit milliseconds is very common. Only use USB/Wifi connection to a computer for certain recording / still image situations - always use the HDMI connection for actual work under the scope. If you wanna record to computer or stream while working, get an HDMI capture card with splitter or passthrough so that your working video feed remains HDMI. Do check monitor reviews if you don't already have a/the screen though, especially gaming hardware reviewers are fairly good at doing latency tests.make sure the test is for HDMI connection. You don't need the best, just trying to avoid any surprise outlier that has a real latency issue. I don't recommend TVs for this, stick with computer monitors.
3: Yes, there is a different learning curve. It will be a bit steeper, especially if you're starting to learn doing work under a microscope at the same time. The depth perception, and acute closeness of the work piece and your hands/tools, feel more intuitive when you're starting out. However, if you were to it that way, going by the anecdotes (!) from those I've discussed this with over the years; more of those who started out working "by monitor" say they find it intuitive and easy, than among those who started out with optics and changed to monitor after several years. None of those who started out with monitor, said or insinuated they had any issues falling back on the optical ports on stereo or trinocular scopes though, even among those who very rarely use the eyepieces. So if you start out learning by monitor, it's not like you need to retrain significantly to move over to optics/the eyepieces.
4: If you already have the trinocular/stimulfocal microscope, then keeping the optics around to have something to fall back on is absolutely a strong recommendation. Especially if others need to use your setup from time to time, and so that you're not unable to work if something suddenly goes wonky with the camera. BUT!
- If you find that you're taking well to working by monitor, you can get a separate C-mount microscope for just the camera. They're *much* more affordable than a trinocular/simulfocal microscope head and mount, and you can even get them in sizes that takes the same barlow lenses as your trinocular scope, so you don't need separate sets of barlows. These microscopes are single port (you've likely seen NorthridgeFix videos, that kind of microscope). With those, the light path is not shared with/split between any other ports like the camera port on a trinocular/simulfocal scope, so you don't need as aggressive lighting, but there are great small format ringlights and coaxial lights for C-mount microscopes for a very reasonable price. The improved light situation is also very favorable if you wanna run a 4k60 cam in 1080p120 mode - through the camera port on a simulfocal trinocular scope, 120+fps *does* quickly quite dark. These microscopes can often provide a bit better working distance, and they drastically reduce how much space around your workpiece is taken up by the microscope and stand. These are much lighter too so articulating mounting arms are more affordable. So if your workbench is against a wall and it's not a long-travel adjustable standing desk - you can easily mount the microscope arm for a C-mount camera microscope on the wall. That way it easily swings away to clear the work area completely when not using the scope, and the little shakes and vibrations of your workbench surface do not transfer and attenuate through the stand/arm fixture as it tends to do when the arm is fixed to the bench/table. The caveat is that you don't mount it to a wall that shakes and vibrates for other reasons of course. And do try to mount it so it screws into wall risers/studs unless walls are concrete ofc.
Don't be dissuaded from trying out the camera and monitor route. Just beware of these core aspects mentioned. A bit steeper learning curve, HDMI straight to decent, low latency monitor, get a C-mount microscope for camera only to free up work space and improve light input to the sensor.
A lot of folks are moving in this direction for good reasons. Good enough equipment for a useable setup has gotten very affordable. Optimum setup if camera and monitor is your primary "input", is a C-mount microscope, but decent 130x-180x units are ~60-70 USD from Eakins so it's not cost prohibitive to try out both if you already have a trinocular scope.