Objective To explore the improvement effect of virtual reality (VR) cognitive training combined with repetitive transcranial magnetic stimulation (rTMS) on cognitive function after stroke, and provide a basis for optimizing rehabilitation plans for post-stroke cognitive impairment (PSCI). Methods A randomized controlled trial (RCT) design was used to include 82 stroke patients who were treated as inpatients or outpatients in the Department of Rehabilitation Medicine of the First Affiliated Hospital of Bengbu Medical University from January 2022 to December 2024, and divide these patients into an observation group (n=41, receiving rTMS combined with VR training) and a control group (n=41, receiving rTMS only). The Montreal Cognitive Assessment (MoCA) Scale, Mini-Mental State Examination (MMSE), and Activities of Daily Living Scale (ADLS) were used to evaluate the changes in cognitive function and life ability before intervention (T0), at 6 weeks (T1), and at 12 weeks (T2). Results The total score of MoCA in the observation group increased significantly compared to the control group (17.25±3.35 → 25.30±3.80 vs 17.15±3.25 → 21.60±3.50) (P<0.05). The improvement of MMSE total score of the observation group was significantly higher than that of the control group (21.55±2.95 → 27.50±2.60 vs 21.45±3.05 → 24.10±2.90) (P<0.05). The total score of ADLS of the observation group improved significantly compared to that of the control group(65.35±6.75→ 79.30±5.80 vs. 64.85±6.95→ 73.80±6.20) (P<0.05). Intergroup comparison showed that the observation group had significantly better improvement in subdimensions such as attention, executive function, and instrumental activities of daily living (IADL) than the control group (all P<0.05). Conclusion VR training combined with rTMS is more effective in improving cognitive function in stroke patients than rTMS alone, providing a new intervention model for post-stroke cognitive rehabilitation.