This study identifies two distinct brain networks involved in task control: the frontoparietal network and the cinguloopercular network. The frontoparietal network, including the dorsolateral prefrontal cortex and intraparietal sulcus, is involved in adaptive control, initiating and adjusting control on a trial-by-trial basis. It emphasizes start-cue and error-related activity. The cinguloopercular network, including the dorsal anterior cingulate/medial superior frontal cortex, anterior insula/frontal operculum, and anterior prefrontal cortex, is involved in stable task set maintenance across the entire task period. These networks operate on different time scales and affect downstream processing through dissociable mechanisms. The study used resting state functional connectivity MRI (rs-fcMRI) data and graph theory to analyze the interactions between these regions. The results showed that the brain's task-control system consists of two distinct networks with different resting state connectivity patterns. The frontoparietal network is associated with adaptive control, while the cinguloopercular network is associated with stable set maintenance. These findings suggest that the brain uses a dual-network model for task control, which better captures the phenomenology of task control than previous unitary models. The frontoparietal network is involved in online control and adjustment, while the cinguloopercular network is involved in maintaining task sets across trials. The two networks are functionally distinct and operate on different time scales. The frontoparietal network is more involved in rapid, adaptive control, while the cinguloopercular network is more involved in stable, long-term control. The study also found that these networks are functionally connected to other brain regions, such as the cerebellum and occipital cortex, which are involved in error detection and sensory processing. The study's findings suggest that the brain uses a dual-network model for task control, which is more resilient and flexible than previous unitary models. This model allows for both adaptive and stable control functions, which are essential for performing a wide range of tasks. The study also highlights the importance of resting state functional connectivity in understanding the brain's task-control system. The results suggest that the brain's task-control system is a complex, adaptive system that can respond to changing task demands and maintain stable control over time.This study identifies two distinct brain networks involved in task control: the frontoparietal network and the cinguloopercular network. The frontoparietal network, including the dorsolateral prefrontal cortex and intraparietal sulcus, is involved in adaptive control, initiating and adjusting control on a trial-by-trial basis. It emphasizes start-cue and error-related activity. The cinguloopercular network, including the dorsal anterior cingulate/medial superior frontal cortex, anterior insula/frontal operculum, and anterior prefrontal cortex, is involved in stable task set maintenance across the entire task period. These networks operate on different time scales and affect downstream processing through dissociable mechanisms. The study used resting state functional connectivity MRI (rs-fcMRI) data and graph theory to analyze the interactions between these regions. The results showed that the brain's task-control system consists of two distinct networks with different resting state connectivity patterns. The frontoparietal network is associated with adaptive control, while the cinguloopercular network is associated with stable set maintenance. These findings suggest that the brain uses a dual-network model for task control, which better captures the phenomenology of task control than previous unitary models. The frontoparietal network is involved in online control and adjustment, while the cinguloopercular network is involved in maintaining task sets across trials. The two networks are functionally distinct and operate on different time scales. The frontoparietal network is more involved in rapid, adaptive control, while the cinguloopercular network is more involved in stable, long-term control. The study also found that these networks are functionally connected to other brain regions, such as the cerebellum and occipital cortex, which are involved in error detection and sensory processing. The study's findings suggest that the brain uses a dual-network model for task control, which is more resilient and flexible than previous unitary models. This model allows for both adaptive and stable control functions, which are essential for performing a wide range of tasks. The study also highlights the importance of resting state functional connectivity in understanding the brain's task-control system. The results suggest that the brain's task-control system is a complex, adaptive system that can respond to changing task demands and maintain stable control over time.