Stroke is an acute cerebrovascular disease in which sudden interruption of blood supply to the brain or rupture of cerebral blood vessels cause damage to brain cells and consequently impair the patient's motor and cognitive abilities. A novel rehabilitation training model integrating brain-computer interface (BCI) and virtual reality (VR) not only promotes the functional activation of brain networks, but also provides immersive and interesting contextual feedback for patients. In this paper, we designed a hand rehabilitation training system integrating multi-sensory stimulation feedback, BCI and VR, which guides patients' motor imaginations through the tasks of the virtual scene, acquires patients' motor intentions, and then carries out human-computer interactions under the virtual scene. At the same time, haptic feedback is incorporated to further increase the patients' proprioceptive sensations, so as to realize the hand function rehabilitation training based on the multi-sensory stimulation feedback of vision, hearing, and haptic senses. In this study, we compared and analyzed the differences in power spectral density of different frequency bands within the EEG signal data before and after the incorporation of haptic feedback, and found that the motor brain area was significantly activated after the incorporation of haptic feedback, and the power spectral density of the motor brain area was significantly increased in the high gamma frequency band. The results of this study indicate that the rehabilitation training of patients with the VR-BCI hand function enhancement rehabilitation system incorporating multi-sensory stimulation can accelerate the two-way facilitation of sensory and motor conduction pathways, thus accelerating the rehabilitation process.