Current clutch simulation, as Scawen said, works well in racing conditions and does its job. The overheating rate is fine imo - for example, I have no problems at all in longer races (I drive with mouse and use kb buttons to change gears and manually operate the clutch), unlike some whining wheel owners
... Thus the clutch is fully raceable, as was said.
However, I think that while the rate of heating is realistic, actual situations when clutch slipping starts are not. As Cue-Ball stated, clutch must be stronger than torque of the engine and grip of the tyres. IRL slipping will occur when "riding" the clutch on take-ofs, hill starts, improperly changing gears - in short, when torque is applied while clutch is partially engaged (or has already been overheated, damaged, worn out). BUT the clutch wont slip (maybe very minimally), when it is fully clamped during take-ofs: IRL when you drop the clutch while reving the engine in say 3rd gear, the engine will immediately bog down and stall, but in LFS revs lower slowly, at the same time slipping and heating the clutch - just try that in XRG. So in this situation the clutch must stall the engine, but in LFS the engine slips the clutch.
This must be caused by two things. First, when engine is peaky or generally not powerful (low torque), it should bog down much faster and stall much easier than now (LX's and formulas moving on idle revs in 2nd gear and not stalling - wrong). Second, when considering powerful/torquey cars (XRT, FZ50), that is, when engine is strong enough not to be stalled/bog down when dropping the clutch in 2nd gear - road tires must lose grip and spin because their grip force will surely be lower that the torque of the engine and the clutch. So, this must be related to longitudinal tyre grip physics - they are a bit too difficult spin and they don't loose enough grip while spinning longitudinally.
My english isn't great, but I hope I explained this quite clear. I understand that Y is finished and it will be left as it is for now. I agree that the current system is raceable and drivable but it doesn't act like IRL in mentioned situations and thus isn't realistic. I just hope that this will be taken into account when engine simulation will be reworked and those longitudinal tyre problems will be fixed (which btw will fix not only standing starts, but also drifting and locked diff behavior imo)
p.s. if somebody doesn't agree on smthng, plz don't instantly flame - wouldn't mind discussing those physics/realism aspects in another thread if needed.