A semantic network is composed of a set of nodes, each representing concepts, ideas, or objects, connected to other nodes. The connections between the nodes, called relations, describe how the concepts are related to each other. Semantic networks are commonly drawn with nodes as labeled ovals connected by labeled arrows, representing relations. Besides simple concept mapping, semantic networks are used for many things such as: natural language understanding, input classification, and automated decision making. Their use has also been prevalent with artificial intelligence for knowledge representation. Traditionally, the relations are predefined and specific to the particular network, and the nodes represent similarly concrete or abstract concepts. Due to this, concepts representing dynamic structures of other relations or relations with distinctive and specific properties cannot be well represented by current semantic networks. This research proposes representing all relations as ordinary nodes, hence, allowing relations to be used just like any other node. This will allow the semantic networks to have: relations that are complex representations (more abstract concepts) of other nodes and relations; relations that can be dynamically related to similar relations; and relations that can be part of structural and inheritance hierarchy, enabling distinct instance relations. The proposed research will enable semantic networks to be a powerful data structure which will extend their current capabilities and allow them to be up to par with current data and computational models.