The term cognitive reserve (CR) is used to describe the lack of direct relationship between the severity of brain damage, or pathology and subsequent levels of observed impairments. It has been suggested by Stern (2009), that CR may reflect differences in (a) pre-existing levels of some reserve capacity of the brain (the passive form); or, (b) differences in the underlying functional architectures supporting cognitive processes (the active form). In this paper, we explore two implementations of cognitive reserve that seek to target both these forms, extending recent work using dynamical systems framework (Baughman & Thomas, 2008; van der Maas et al., 2006). We examine how variability in cognitive reserve may modulate the effects of damage, at different levels of intelligence. The resulting simulations showed that level of intelligence does not differentially modulate the pattern of cognitive change following complete destruction of a single cognitive process, but that the effects of damage are proportionate across each level of intelligence. Following the two implementations of cognitive reserve that we tested, we found: (1) higher levels of connectivity within a given architecture resulted in greater spread of damage and lower endstate performance; and, (2) functional architectures that are characterized by greater specialization of function, rather than distributed function, differentially protected against the effects of damage, with these models also exhibiting better recovery.