The contemporary data landscape, characterized by diverse data types and escalating volumes, has driven a significant evolution in database management systems. Traditional Relational Database Management Systems (RDBMS) often struggle with the scalability and flexibility required for modern applications, while NoSQL databases, though scalable, frequently compromise on transactional consistency. NewSQL systems have emerged as a hybrid solution, aiming to combine the ACID guarantees of RDBMS with the horizontal scalability of NoSQL. This report explores the critical need for multi-model capabilities within NewSQL DBMSs to efficiently manage heterogeneous data adhering to various data models—including relational, document, graph, and key-value—within a single, unified database. It delves into the architectural considerations for supporting such diversity, examining storage paradigms like row-store, column-store, and hybrid approaches, alongside data distribution strategies such as sharding and partitioning. Furthermore, the paper investigates advancements in multi-model query languages, particularly SQL++ and extensions in SQL:2016, and discusses query optimization techniques essential for handling complex, hybrid workloads. Finally, it addresses the performance evaluation of these systems in big data environments, highlights current limitations, and outlines future research directions for achieving truly versatile and high-performing data management solutions.