The National Weather Radar Testbed (NWRT) located in Norman, Oklahoma was established to demonstrate the potential to simultaneously perform aircraft tracking, wind profiling, and weather surveillance as a multi-mission phased-array radar (MPAR). Since its inception in September of 2003, the system has undergone an extensive engineering evaluation and numerous hardware and software upgrades. However, in spite of significant engineering work, the real-time signal processing functionality currently implemented in the PAR is limited. Even with these limitations, several research experiments have successfully demonstrated many of the unique advantages of using phase-array technology in the context of weather observation. A modern and improved multi-processor/multi-computer signal processing environment will allow the implementation of new and advanced real-time signal processing techniques. These include schemes to effectively remove clutter contamination from meteorological signals, methods to mitigate range and velocity ambiguities, and techniques that allow for faster data collection. This paper presents initial results and describes the roadmap of planned signal processing upgrades for the NWRT that will provide researchers and users with an optimum platform for demonstrating and evaluating the MPAR concept.