Oh dear.
Quite apart from the outdated geometry assumptions and data used in the VTech paper (also apparent in their weight estimates, as discussed here a couple of months ago), you are completely neglecting engine thrust degradation with altitude due to the drop in air density.
AL-31F dry thrust (at sea level, zero speed, ISA conditions) is not ~39klb for two engines, but
only about 34klb (2x76kN) - and then only on the test stand:
installed in the aircraft, it's not even 33klb.
View attachment 48266
According to the drag graph,
Mach 1.1 @ 35000 feet requires >14klb of thrust - but in those conditions a pair of AL-31Fs (presumably with Su-27 inlets, though at such low supersonic Mach the design hardly matters)
delivers just 60kN/13.5klb. If the various errors in the VTech analysis turn out favourably, this is close enough that Mach 1.1 might be doable on dry thrust - but only if the J-20 punches through the sound barrier in reheat.
View attachment 48267
Of course, for all we know the engine might be an advanced Salut-developed derivative with capabilities more like the 117S (which would in fact hit 39klb ISA SLS), however that needless to say is subject to similar thrust degradation. It might push speed out to Mach 1.3, but almost certainly no more than that (and it'd likely still be incapable of accelerating up to that point without afterburner).