We previously reported that the phosphodiesterase-5 (PDE5) inhibitor sildenafil prevented neurodegeneration but not learning deficits in middle-aged rats that were subjected to the permanent, three-stage, four-vessel occlusion/internal carotid artery (4-VO/ICA) model of chronic cerebral hypoperfusion (CCH). In the present study, we examined whether the PDE3 inhibitor cilostazol alleviates the loss of long-term memory (i.e., retrograde amnesia) caused by CCH. The effect of sildenafil was then compared to cilostazol. Naive rats (12–15 months old) were trained in a non-food-rewarded eight-arm radial maze and subjected to CCH. One week later, retrograde memory was assessed for 5 weeks. Cilostazol (50 mg/kg, p.o.) was administered for 42 days or 15 days, beginning approximately 45 min after the first occlusion stage. Sildenafil (3 mg/kg, p.o.) was similarly administered for 15 days only. Histological examination was performed after behavioral testing. Chronic cerebral hypoperfusion caused persistent retrograde amnesia, which was reversed by cilostazol after both short-term and long-term treatment. This antiamnesic effect of cilostazol was sustained throughout the experiment, even after discontinuing treatment (15-day treatment group). This effect occurred in the absence of neuronal rescue. Sildenafil failed to prevent CCH-induced retrograde amnesia, but it reduced hippocampal cell death. Extending previous findings from this laboratory, we conclude that sildenafil does not afford memory recovery after CCH, despite its neuroprotective effect. In contrast, cilostazol abolished CCH-induced retrograde amnesia, an effect that may not depend on histological neuroprotection. The present data suggest that cilostazol but not sildenafil represents a potential strategy for the treatment of cognitive sequelae associated with CCH.