The issue here is recovery from very low RRPM and it is a fact that the autorotative force (lift) on the disc is dependant on the angle of attack and velocity (rpm) squared, irrespective of helicopter type.
It is an aerodynamic fact that if you have low RRPM and forward cyclic is applied, the angle of attack is decreased, leading to less autorotative forceand more drag. If autorotative force is insufficient to overcome drag then the RRPM will decrease and the blade will stall which is what is suggested happened to the helicopters in the article.
Forward cyclic also unloads the disc which, as discussed above tends to lower RRPM. not good.
Aft cyclic is not about controlling attitue, speed or anything else at this time. It is about controlling inflow angle to give maximum autorotational force. RRPM is the only thing in life just now!
Tight turns load the disc and will cause increase in RRPM which crucially increases the efficiency of the autorotative 'engine' (remember the RPM squared bit) so conservation of angular momentum does apply but the increased autorotative force will more than offset this and on rolling out of the turn the RPM will not fall as much as expected. Agreed the disc does not know it is in a turn but it is working much better at higher RPM.
HOWEVER:
If RPM is low, increasing the inflow angle by any of these means may well cause more of the blade to stall if the critical angle is exceeded. If enough of the blade stalls, the drag goes massive and the RPM will fall further. In other words the disc will stall at a higher RRPM.
I can't find much info on recovery from very low RRPM but aerodynamically, a flare or tight turn seems not too bad an idea. If it causes blade stall you are probably really in trouble and will never know if not flaring was a better option.
better still, don't go there!
