Effectively thermal conduction pathways are essential for achieving high thermal conductivity (TC) in polymer-based composites. While constructing a segregated structure can yield high TC with low filler loadings, traditional processing methods often have inherent drawbacks, typically involving a complex preparation process and reduced mechanical performance. In this study, a free-form was constructed based on the full advantages of 3D printing. The composite with a dense segregated structure was prepared by 3D printing a scaffold followed by filling with a thermally conductive filler and then hot pressing. Furthermore, boron nitride (BN) was chosen as the electrical insulating and thermally conductive filler model, while graphene (GR) served as the electrical and thermally conductive filler model. Benefiting from the interconnected thermal conductivity network, the 3D-printed composites with GR exhibit a high thermal conductivity of 3.82 W/mK, representing 2.81 times that of composites with randomly blended fillers. Concurrently, these composites also demonstrated robust mechanical properties, achieving tensile strengths of up to 20.3 and 40.6 MPa, respectively, signifying substantial improvements of 1.83 and 3.98 times over their randomly blended counterparts. Therefore, this strategy provides the way for the easy, effective, and universal preparation of thermally conductive composites suitable for various insulated or electrical electronic devices.