Probabilistic Boolean Networks (PBNs), a modeling tool in bioinformatics for Gene Regulatory Networks, are capable of capturing the dynamics of complex biological systems and non-biological systems, including manufacturing systems, machinery, and smart grids. In this proof-of-concept study, we introduce a PBN model designed to emulate state maintenance in a manner akin to Artificial Neural Networks, suggesting a potential alternative to such networks. This concept was evaluated through a PBN model of an Intelligent Power Router, a device proposed for efficient power management within power generation, transmission, and distribution systems. The model successfully learned equality and negation functions across different experiments. By leveraging the intricate properties of PBNs, we demonstrate their potential as adaptive learning tools with positive feedback and propose that these networks may have broader applications than previously recognized.