Background and Objectives: Osteoarthritis (OA) of the knee and hip is a major cause of pain and functional impairment. This study evaluated the effects of non-immersive virtual reality (NIVR) combined with conventional physical therapy (CPT) on pain intensity, mechanical hyperalgesia, and perceived recovery in older adults with OA. Materials and Methods: Sixty older adults with mild-to-moderate knee or hip OA were randomly assigned to a NIVR group (NIVR-G; n = 30) or a CPT group (CPT-G; n = 30). Both groups completed 30 sessions over 10 weeks (3 sessions/week). The NIVR-G performed 20 minutes of exergames integrated into CPT. Pain intensity was assessed using the Visual Analog Scale (VAS), and mechanical hyperalgesia was evaluated through pressure pain thresholds (PPTs). Secondary outcomes included the Global Rating of Change (GRoC) and the minimal clinically important difference (MCID) for the VAS. This study is a secondary analysis of a randomized controlled trial registered at ClinicalTrials.gov (ID: NCT05839262). Results: The NIVR-G demonstrated significant reductions in pain intensity after 30 sessions (p < 0.05, d = 1.50), with greater improvements compared to the CPT-G (p < 0.05, d = 1.17). The MCID for the VAS was established at 9.2 mm, with a higher proportion of responders in the NIVR-G (p < 0.05). The NIVR-G also reported superior recovery perception on the GRoC scale (p < 0.05). No significant changes in PPTs were observed in either group. However, the improvements in the NIVR-G diminished four weeks post-intervention. Conclusions: NIVR exergames combined with CPT significantly reduced pain intensity, improved perceived recovery, and resulted in a higher proportion of responders compared to CPT alone. These findings support the use of NIVR as an effective adjunct to CPT in older adults with OA; however, further research is needed to optimize its long-term benefits.