Like other posters, I have found the behavior of the Cup car very 'bouncy' and with quite abrupt FFB traits, and I'm not sure the real car reacts in this way. I don't think this is just down to FFB tweaking - perhaps the nature of this car is to create very high steering rack forces compared to other 'more pure' racing cars, so we have to adapt to suit; but I'd be more of the opinion this is a physics issue with the car, and the forces on the steering rack feel exaggerated.
As regards the setup, the first thing I did as well was to reduce the toe at the rear and also to reduce the camber at both ends significantly. I've not gone as far as some yet. I also reduced the tire pressures quite a lot, which clearly made a great difference at Sebring for instance in overall grip both under heavy braking areas and traction out of those slower corners. The baseline setup seems awfully extreme.
There's a tendency in all sims to have 'minimum tire pressures/minimum ride height/maximum camber' be the fastest, and without much effect on long-term running of the tire. In my opinion, this shouldn't be right. Even if faster on a single lap, handling should go away quickly and average lap times should be much better with logical pressures/cambers for long run results. In fact, lowering the tire pressures always seems to have a positive result, and that's wrong. The temperatures that build up in a tire are a result of friction on the tire surface (sliding) but most importantly to the heating effect of constant stretching (like what happens when you do this to a rubber band, the heat buildup is huge). Further, this heat escapes via two methods: convection to air, and conduction to the ground (which is why track surfaces heat up) and again its mostly due to the latter not the former. I think most people don't understand these. In fact, when we listen to what happens in most series (F1, IMSA) when the tires start running too hot and there is a chance of failure of the tire, the instructions are always to run a minimum pressure higher than teams might want to go for outright performance. In other words, increasing the tire pressures results in LOWERING the temperatures in critical regions of the tire. We don't seem to have this in any sim-tires, yet. Real world, either the temps get too high and you melt the beads (keeps tire on the rim) or it melts the shoulder area and you get delamination of the tread from the sidewalls. (I have experience of this working at Cooper Tire when there was the famous Ford Explorer Firestone tire failure debacle). Without the penalties of these types of failures, or the existence of critical regions, it seems the answer is to continually lower pressures (or the only way to avoid it is set minimum allowed garage pressures in the car files). iRacing have overdone the temperature results and grip reductions in the past, which I think is a direct result of this, and their latest NTMv7 tires are supposedly now trying to bring in concept of failures so setups aren't so extreme.