6G, as the next generation of mobile communication networks, should provide a universal platform for a wide variety of future connected applications with diverse requirements in terms of throughput, latency, reliability, and more. Additionally, with an exponentially growing number of devices and generated data, 6G needs to be much more sustainable and energy-efficient than previous generations, requiring efficient orchestration and optimization of resources on all layers of the network. The ability to quickly self-adapt to changing environmental conditions and application behaviors becomes a crucial aspect of a performant and efficient network operation. While current software-based implementations of self-adaptivity provide high flexibility and the potential for complex optimizations, they often suffer from communication and computational overhead, limiting their responsiveness, and can only adapt to what the underlying hardware offers. This necessitates an overhaul of current self-adaptive mechanisms in software and especially in hardware. The X-MAPE concept proposes an extension of self-adaptive algorithms with predefined, configurable, Reflexive actions, providing an immediate path between monitoring a specific change and executing an appropriate action, allowing faster reactions. It further enables adaptivity on layers that were previously not considered, covering the whole network architecture from devices like cyber-physical systems to network equipment and computing nodes. This chapter presents a vision for adaptive computing systems for future communication networks, the abstract concept of Reflexes in self-adaptive systems, and several realizations of this principle in various components across the layers of the network infrastructure.