Background: Survival benefits of self-reported recreational physical activity (PA) during cancer survivorship are well-documented in common cancer types, yet there are limited data on the associations between accelerometer-derived PA of all domains, sedentary behavior, and mortality in large, diverse cohorts of cancer survivors. Methods: Participants included adults who reported a cancer diagnosis in the National Health and Nutrition Examination Survey and wore an accelerometer for up to 7 days in 2003-2006. Participants were followed for subsequent mortality through 2015. We examined the association of light PA, moderate to vigorous PA, total PA, and sedentary behavior, with all-cause mortality. Cox proportional hazards models estimated hazard ratios (HRs) and 95% confidence intervals (CIs), adjusting for demographics and health indicators. Results: A total of 480 participants (mean age of 68.8 years [SD = 12.4] at the time of National Health and Nutrition Examination Survey assessment) reported a history of cancer. A total of 215 deaths occurred over the follow-up period. For every 1-h/d increase in light PA and moderate to vigorous PA (MVPA), cancer survivors had 49% (HR = 0.51, 95% CI = 0.34 to 0.76) and 37% (HR = 0.63, 95% CI = 0.40 to 0.99) lower hazards of all-cause mortality, respectively. Total PA demonstrated similar associations with statistically significantly lower hazards of death for each additional hour per day (HR = 0.68, 95% CI = 0.54 to 0.85), as did every metabolic equivalents of task-hour per day increase in total PA estimations of energy expenditure (HR = 0.88, 95% CI = 0.82 to 0.95). Conversely, more sedentary time (1 h/d) was not associated with statistically significantly higher hazards (HR = 1.08, 95% CI = 0.94 to 1.23). Conclusions: These findings reinforce the current recommendations for cancer survivors to be physically active and underscore the continued need for widespread PA promotion for long-term survival in older cancer survivors.