Dear Editors, I am pleased to submit our extended manuscript entitled "Unlocking the Potential of Generative AI through Neuro-Symbolic Architectures – Benefits and Limitations" for consideration in the NeuroSymbolic AI Journal special issue on NeSy 2025 Extended Papers. This extended version builds upon our original NeSy 2025 conference contribution by offering a more in-depth and structured analysis of the neuro-symbolic AI landscape. First, we present a comprehensive classification of neuro-symbolic architectures. Each category is defined based on the directionality of information flow, the level of coupling between neural and symbolic components, and their functional roles within a hybrid system. Second, we propose a novel mapping between generative AI technologies and these neuro-symbolic architectures. For each class of architecture, we identify relevant generative approaches—such as large language models, retrieval-augmented generation, graph neural networks, variational autoencoders, and multi-agent systems—and describe how they can be effectively integrated with symbolic reasoning modules. This mapping illustrates the compatibility and complementarity of these technologies across different NSAI design patterns. Third, the paper introduces an evaluation framework based on seven core criteria: generalization, scalability, data efficiency, reasoning ability, robustness, transferability, and interpretability. We assess how each architectural category performs across these dimensions, helping researchers and practitioners make informed decisions when selecting or designing a hybrid model for a given use case. Finally, we explore how these neuro-symbolic architectures can be applied in the field of 4D printing. We highlight specific scenarios—such as the inverse design of smart structures, optimization of multi-material configurations, control of shape transformations, and multi-scale modeling—where the hybridization of symbolic logic and generative models can offer significant advantages in terms of intelligence, adaptability, and automation. Thank you for considering our submission. We look forward to your feedback and the opportunity to contribute to this special issue. Sincerely, Oualid Bougzime (corresponding and first author) ICB UMR 6303 CNRS, Université Marie et Louis Pasteur, UTBM Email: oualid.bougzime@utbm.fr