Making engineering scale models is much more tricky than usually assumed, because dimensions do not all scale the same.
As an example, how do I make a scale model of a pillar? If a scale, say, every dimension of the pillar to one tenth of the original, the weight of the pillar will be 1/1000 of the original (10 cubed), but the section of the pillar will be only 1/100 of the original (10 squared). So every square meter of the pillar section will only support 1/10th of the original weight, and if you want to study the ability of the pillar to sustain its own weight you will need to use a material which is only 1/10th as resistant (or 1/10th as dense), which is usually not an easy thing to do (*).
Now add in some more variables you can be interested in, say potential energy, which scales with the 4th power of the linear dimension (three powers for the mass, one more for the height) and the modelling quickly becomes an inextricable mess, which at best is a serious work for specialized engineers and where often the different constraints become impossible to satisfy at the same time (**). So, no wonder a scale model of WTC can have problems in replicating the collapse. At the very least, building it would require a serious engineering effort, if it's possible at all.
(*) By the way, this also explain why the usually hyped claim 'ants can lift 100 times their boidy weight' is, actually, a triviality.
(**) Off rtopic, but one of my favourite examples of a scale model spectacularly failing is the 1:200 model of the Vajont basin and dam, in northern Italy, which was made in 1961 to study the possible effects of an avalanche falling from Mount Toc into the Vajont artificial lake. It was a pretty big model:
Di VENET01 - Opera propria, CC BY-SA 4.0,
https://commons.wikimedia.org/w/index.php?curid=55348340
https://it.wikipedia.org/wiki/Disas...el_modello_idraulico_del_serbatoio_del_Vajont
and it was made by competent (even if wishful thinking) engineers. After one year of experiments they concluded that there were no risks, provided the lake level would be lowered to a 'safe' height. The lake level had just been lowered to the prescribed height when, on October 9, 1963 Mount Toc gave way and a big avalanche fell into the lake. The resulting wave from the 'safe', according to the model, lake level overtopped the dam by (it's been estimated) something like 250 meters, resulting in a mega-tsunami which claimed the life of (officially, many corpses, buried under meters of mud, have never been found) 1917 people.
https://en.wikipedia.org/wiki/Vajont_Dam