Cognitive dysfunction and negative symptoms of schizophrenia remain an unmet clinical need. Therefore, it is essential that new treatments and approaches are developed to recover the cognitive and social impairments that are seen in patients with schizophrenia. These may only be discovered through the use of carefully validated, aetiologically relevant and translational animal models. With recent renewed interest in the neurodevelopmental hypothesis of schizophrenia, postnatal administration of N-methyl-D-aspartate receptor (NMDAR) antagonists such as phencyclidine (PCP) has been proposed as a model that can mimic aspects of schizophrenia pathophysiology. The purpose of the current review is to examine the validity of this model and compare it with the adult subchronic PCP model. We review the ability of postnatal PCP administration to produce behaviours (specifically cognitive deficits) and neuropathology of relevance to schizophrenia and their subsequent reversal by pharmacological treatments. We review studies investigating effects of postnatal PCP on cognitive domains in schizophrenia in rats. Morris water maze and delayed spontaneous alternation tasks have been used for working memory, attentional set-shifting for executive function, social novelty discrimination for selective attention and prepulse inhibition of acoustic startle for sensorimotor gating. In addition, we review studies on locomotor activity and neuropathology. We also include two studies using dual hit models incorporating postnatal PCP and two studies on social behaviour deficits following postnatal PCP. Overall, the evidence we provide supports the use of postnatal PCP to model cognitive and neuropathological disturbances of relevance to schizophrenia. To date, there is a lack of evidence to support a significant advantage of postnatal PCP over the adult subchronic PCP model and full advantage has not been taken of its neurodevelopmental component. When thoroughly characterised, it is likely that it will provide a useful neurodevelopmental model to complement other models such as maternal immune activation, particularly when combined with other manipulations to produce dual or triple hit models. However, the developmental trajectory of behavioural and neuropathological changes induced by postnatal PCP and their relevance to schizophrenia must be carefully mapped out. Overall, we support further development of dual (or triple) hit models incorporating genetic, neurodevelopmental and appropriate environmental elements in the search for more aetiologically valid animal models of schizophrenia and neurodevelopmental disorders (NDDs).