Despite its significance for health and education, the neurocognitive mechanism of real-life self-control remains unclear. While recent studies focused on task-related brain activation patterns as predictors of self-control, the contribution and relevance of functional connectivity between large-scale brain networks mediating higher-order cognition is largely unknown. Using a saliency-based triple-network model of cognitive control, we tested the hypothesis that cross-network interactions among the salience network (SN), the central executive network (CEN), and the default mode network (DMN) are associated with real-life self-control. To this end, a large community sample (N = 294) underwent ecological momentary assessment of daily self-control as well as task-free fMRI to examine intrinsic inter-network organization and determine a SN-centered network interaction index (NII). Logistic multilevel regression analysis showed that higher NII scores were associated with increased real-life self-control. This suggests that the assumed role of the SN in initiating switching between the DMN and CEN is an important part of self-control.