In recent years, several pathophysiological models of schizophrenia, i.e. the early and late brain neurodevelopmental and post-illness onset neurodegenerative models, have been proposed and theorists have often argued as if these explanations are mutually exclusive. We propose that all these mechanisms may interact cumulatively during successive critical 'windows of vulnerability' during brain development and during the early course of the illness to lead to the clinical manifestations of the illness. Early brain insults may lead to dysplasia of selective neural networks that account for the premorbid cognitive and psychosocial dysfunction seen in many patients. The onset of psychosis in adolescence may be related to an excessive elimination of synapses and secondarily, phasic dopaminergic overactivity. Following illness onset, these neurochemical alterations in relation to continuing untreated psychosis may lead to further neurodegenerative processes. A reduction in tonic glutamatergic neurotransmission and a phasic glutamatergic excess can potentially predispose to these processes and may have considerable explanatory power. This hypothesis is consistent with central characteristics of schizophrenia such as premorbid manifestations, adolescent onset, functional decline early in this illness, cognitive impairments, the role of dopamine and the role of genes and environment in pathophysiology. This 'three hit' model extends similar integrative conceptualization by other investigators and generates testable predictions of relevance to future pathophysiology and treatment research in schizophrenia.