Abstract: Background/Objectives: Extensive evidence suggests that exercise is physically and mentally beneficial for cancer survivors. This study reports on changes in self-reported stress, physiological biomarkers for stress (salivary cortisol), and HR-QOL constructs for fifty breast cancer survivors participating in one of three different exercise programs over 6 months. Methods: Fifty post-treatment breast cancer survivors were randomized to either therapeutic yoga-based exercise (YE), comprehensive exercise (CE) (aerobic, resistance, flexibility), or choosing (C) their own exercise activities. Participants completed the Perceived Stress Scale (PSS), Medical Outcomes Short-Form 36® (SF-36), and the Pittsburgh Sleep Quality Index (PSQI). Five samples of salivary cortisol were collected on two consecutive days. The 10 samples were used to calculate the diurnal rhythm slope. Outcome measures were repeated after six months. Results: All groups improved in HR-QOL measures of PSS; PSQI sleep quality components of latency and daytime functioning; and five of the ten SF-36 scales (Mental Component Scale, Social Functioning subscale, Mental Health subscale, Physical Component Scale, Physical Functioning subscale). Although the CE group observed the most favorable change in cortisol (−0.183), where cortisol slope changes approached significance (p = 0.057), but no significant decrease in cortisol between groups were noted. Conclusion: Our results suggest that it is the engagement of, rather than the specific type of exercise, which is associated with improved HR-QOL. However, longer-term studies with better adherence monitoring and larger sample sizes are needed to better determine clinical impact.