Efficient and durable operation of proton exchange membrane electrolyzers (PEMELs) requires a detailed understanding of the coupled electrochemical, thermal, and transport phenomena that govern hydrogen production and component degradation. Conventional monitoring approaches, which rely primarily on physical sensors and empirical correlations, are insufficient to capture internal states that critically influence efficiency, durability, and dynamic performance. To address these limitations, this work presents an advanced, computationally efficient, physico-chemically consistent modelling basis for use in a State-of-X (SoX) framework tailored to PEMEL. At the core of the proposed framework is a reduced-order, multi-scale PEMEL model that resolves the dominant physical processes across the membrane–electrode-assembly and flow channels while remaining suitable for faster-than-real-time execution. The model incorporates detailed sub-models for electro-osmotic and pressure-driven membrane water transport, gas crossover, two-phase transport in porous transport layers, and mixed-potential effects under dynamic operating conditions. Particular attention is given to accurately capturing the strong coupling between submodels, which is essential for both performance prediction and degradation diagnostics. The modelling framework is designed to support State-of-Operation-Conditions (SoOC) and State-of-Health (SoH) diagnostics through systematic identifiacation of parameters and observation of internal states of the model rather than extensive sensor deployment. Internal states such as local gas saturation, reaction overpotentials, and current density distributions are reconstructed using only boundary conditions and global electrical measurements. Long-term changes in identified model parameters are directly linked to degradation mechanisms, including catalyst layer loss, and increased gas crossover. Numerical analyses and experimental validation show that the proposed model-based SoX framework reliably captures both transient and steady-state PEMEL behaviour across a wide operating envelope. By enabling continuous consistency between model predictions and measured system responses, the approach provides a robust foundation for advanced control, diagnostics, and lifetime optimisation of PEMEL, supporting their large-scale deployment in renewable hydrogen systems.