Heterogeneous interacting socio-technical systems, distinct by their own cultures, structures, methods, and machines, are transforming the process of making informed decisions and assessment of situations in the context of cybersecurity into a serious challenge. Stakeholders in each system are forced to take actions at any given time based on the situation. These actions not only affect their own, but also the whole system's properties. Therefore, maintaining cybersecurity will become more difficult. In this work in progress, we attempt to apply a cognitive learning model in agent-based computational economics to create a comprehensive environment for cybersecurity situation modeling, reasoning and learning. Taken together, this enables us to holistically assess the situations and systems' reflexivity, and support human and/or computational understanding of the situations. The aim of this study is to design and develop a hands-on cybersecurity platform to train cognitive situation management in cyber ranges.