How can a power converter small signal model predict transient behaviour during a load step which is a large signal disturbance?

I see small signal model being used to predict the Overshoot/Undershoot of output voltage due to a load step.

Image may be NSFW.
Clik here to view.

How do we justify this approach, since the basic assumption of small signal modelling is the disturbance is small wrt DC values?Image may be NSFW.
Clik here to view.

During the load step,$\Delta$ Iout would typically be large.