I actually did skim the report. I'm not a structural engineer, so I'm not entirely clear on all of the terminology and how reasonable some of the decisions are, but here's the highlights of what I recall.
* The NIST report made some assumptions that are not justified by actual evidence. The critical complaints are about the modelling of the exterior wall as a unit (?) that made it unreasonably stiff, and insufficient modelling of the actual joints between structural members. [I can't speak as to this accuracy, this is well beyond my understanding.]
* They did their own model of how the structure would behave due to heat and cannot confirm the NIST results.
* They also modelled the collapse on their own. I'm not 100% certain how the model works, but it looks like they basically deleted key structural members and observed how the structure would collapse when they did so.
* In order to match the recorded evidence, they had to delete key structural members very high up (higher than any floors known to have been on fire) to mirror the observed "penthouse kink", and they had to delete all the columns on an entire floor to mirror the fall. (And of course, no plausible mechanisms for either to occur other than controlled demolition).
If I'm following the report correctly--which is a big if--then they have two separate models which aren't used in conjunction with one another, and the first one "overturns" the analysis on how the fire could have caused a key structural member to fail, and the other one "concludes" that the key structural member failing couldn't have caused the collapse. In the latter model, as I understand it, there's no attempt to account for the deteriorated state of the building prior to the final collapse, for example the large gash in the building known to have occurred as a result of falling debris, or any putative (but not confirmed) damage that might have been caused by the long-running fire.
Overall, I didn't come away feeling impressed by their analysis.