Wow. Well done.
That said, there are a few issues with "simplify," since the baseline is not simple (existing Orion, launched by SLS), causing literally everything else to be less simple.
If the entire system had been designed from the start with a mission, they would not be having this issue. Hence the usual complaint (at least from me), that SLS is "a rocket to nowhere."
So we have a bad, required starting point (Orion), and added to that is an entirely different mission from Apollo (here's a link to the PDF he mentioned). Apollo was brief duration stays. Artemis is meant to be longer stays, both to be less rushed, and ALSO because Orion/Gateway won't be back for at least 6.5 days, so that's kind of the minimum stay. If the mission was changed to mimic the simplicity of Apollo within the Orion requirements, then you still need something like the LM, but fit for at least 6.5 days habitation.
So take a specific observation he makes about hypergolics. He's right of course, but then you need a lot of hypergolics. The lander must be bigger for the required mission duration.
A lander that is 20t dry needs to be ~90t wet to do the RT from NRHO using methalox. Using hypergols? It needs to be 118t wet.
What about a 10t lander (~2X the Apollo LM)? 45t with methalox, 59t with hypergolics.
So that gives us some sense of what the simple solution could look like. A ~60t (wet) lander done with hypergolic props. Except you need a way to send a 60t lander to Gateway/NRHO.
Interestingly, Starship, assuming a 75t variant (which is actually pretty beefy, could be lower), could send the 60t lander to NRHO (I'm having it fly all the way to Gateway here, else lander needs more dv) with 225t of residual props. That's 285t of payload, including lander. In short, and expended SS could send a 10t dry hypergolic lander to Gateway. Note I assumed a single stage lander.