Let's get back to basics.
An aerodynamic stall is when airflow separates from the wing. It goes from laminar flow to turbulent. The point of separation moves from the back of the wing to the front as AOA increases. The stall is the point where the weight of the aircraft (if in level flight) exceeds the lift generated.
The airflow separation reduces the effective area of the wing and reduces lift.
An accelerated stall is when the load on the wing increases to exceed the lift on the wing.
Recovery of a level stall is to lower angle of attack. Pushing the nose over does this, as well as increasing speed and unloading the wing.
Recovery from an accelerated stall is simply reducing the load on the wing by reducing the G's.
Many factors influence the exact characteristics of a stall:
Wing form.
Yaw angle.
Wing condition. Planes are often slightly bent.
Engine torque. I could never get a Cessna 152 to spin against torque.
My point is that modeling the chaotic physics of a stalling aircraft is difficult, if not impossible. Computational Fluid Dynamics is a science all its own. To keep the simulation running at a playable speed, I assume HiTech does some educated approximations of how the planes will depart controlled flight.
*** I post this, because I could not follow or was unfamiliar with some of the theories expressed in previous posts. This is my knowledge, based on my time as an instructor pilot in the USAF.
***