<p>Digital Twin (DT) technology, a powerful technology that creates virtual representations of physical entities, has rapidly expanded across multiple sectors, enabling real-time data integration, predictive analytics, and the optimization of complex systems. However, existing reviews often focus on narrow application domains, lacking a combined perspective on modeling, integration and emerging technologie<b>s.</b> This paper provides a comprehensive review of DTs’ current challenges and future directions, focusing on various facets, such as high fidelity and order reduced surrogate models, probabilistic modelling, data lifecycle management and cyber security. This review bridges the gap by synthesizing advancements across modeling techniques, integration with artificial intelligence (AI), Internet of Things (IoT), Blockchain, and cross-sector applications, offering a holistic view of DT development. This paper examines the technical and practical limitations of adopting DT, including scalability, data privacy, interoperability, and integrating emerging technologies, such as artificial intelligence, blockchain, and machine learning. It discusses the application of DT in diverse fields, such as manufacturing, healthcare, urban management, energy grids, and 6G networks. Findings indicate that despite significant advancements, substantial obstacles remain in modelling complex, data-intensive systems, securing real-time communication, and addressing ethical and regulatory challenges. This review offers insights into the key areas where DT can evolve, emphasizing the need for improved data models, robust security frameworks, and interdisciplinary collaboration to maximize the potential.</p>