Can virtual feedback training improve performance in the physical world? Can enactment enhance this feedback training effect? This study addresses these questions using two affordance judgment (AJ) tasks, implemented in both physical (PE) and virtual (VE) environments. Sixty-four young adults completed two tasks. The Aperture Task (Oculus Rift head-mounted display, HMD) was included to test replication of our prior findings from Gölz et al. (Virtual Real 27:1697–1715, 2023). We further wanted to test whether feedback training would also be effective in a different type of AJ task, a Reachability Task (Meta Quest 2 HMD). The procedure included a virtual feedback training with or without producing a simple forward arm movement (occluded enactment). AJ performance was assessed using accuracy, perceptual sensitivity, and response bias (judgment tendency). After feedback training in the VE, participants demonstrated significant improvements across all three measures in both tasks within the VE. In the Aperture Task, improvement from pre- to post- in the PE did not reach significance (replication). However, a significant feedback training transfer was measured for the Reachability Task. In this study, no advantage emerged for feedback training with enactment versus feedback training without enactment. Our findings highlight the potential and current limitations of virtual settings for feedback training of AJs.