Industry 4.0 has transformed automation in production systems and processes through customized and intelligent technologies, and laid the technological, data, and communication protocol foundation to propel human-centric Perceptive Disassembly Systems, under an Industry 5.0 paradigm. Disassembly, the initial stage of circular economy reverse cycles, involves dismantling end-of-use (EOU) products to retrieve high-value components for repair, remanufacturing, and recycling. Efficiency in disassembly systems is hindered by challenges such as uncertain product quality, quantity, batches, and timing. In addition to EOU product variability, automating disassembly is challenging for companies due to budget limitations, upfront costs, and the need for qualified personnel to maintain operations. This paper proposes a novel human-centric disassembly paradigm, a Perceptive Disassembly System that integrates extended reality, collaborative robots, and dynamic data driven application systems to enable seamless collaboration between humans and machines. The system aims to empower disassembly workers by providing real-time assistance and enhancing cognitive automation during disassembly tasks. By leveraging knowledge modeling, real-time environment modeling, user modeling, and dynamic recommendation systems, the Perceptive Disassembly System seeks to transform traditional disassembly processes into intelligent, human-centric workflows. The contribution of this paper lies in its innovative approach to addressing the challenges of uncertainty in disassembly environments through the modeling, integration, and implementation of a real-time collaborative and augmented disassembly system. By bridging the gap between human cognition, task modeling, and automation, this research has the potential to revolutionize disassembly operations, paving the way for more efficient and sustainable recovery practices.