Most surprisingly, it seems there's a formal paradox in the thermodynamics of self-gravitating systems. As Piet Hut (of Barnes-Hut fame) at the IAS says:
"A thermodynamic treatment of self-gravitating systems is fraught with peril: from a formal point of view, it cannot even be defined, because there is no thermodynamic limit." - Hut, Gravitational Thermodynamics [1]
Also surprising, thanks to JWST, we can now see clearly into the center of the Milky Way, and there's CO ice![2] Probably more types, but JSWT doesn't have the instrument for this.
This is related because that means it's colder than expected, but also quite full of gas (that was obscuring our view until JWST and near microwave). But cold dense gas should collapse to form stars, and we're not seeing nearly what we should.
Both suggest we're missing something about the grav/thermo relationship, and what's called the Jeans Mass[3]. If you go digging, you'll find our Sun should have taken a planetary nebula about 20x larger than what we have. This in turns suggests that simple static models are maybe misleading, as there's usually dynamics almost everywhere we see large gas clouds. And then this in turn begs the question of why gravitational collapse is needed if there's already large kinetic flows involved, slamming mass together.
Anyways, I'm a total amateur and quite possibly have all this wrong. Just dropping in case anyone's interested!
[1] https://www.ias.edu/ids/~piet/act/phys/thermo