Fisheries stock assessment is an important prerequisite for formulating sustainable fishery development strategies. Limited data is a common challenge to global fishery stock assessment, with traditional stock assessment methods inapplicable to the stock assessment of data-limited fisheries, because of the need for reliable biological information for accurately estimating species abundance. Owing to limited funding and large datasets required in such surveys, only 1% of fish stocks have had systematic assessments conducted. Therefore, it is difficult to assess the maximum sustainable yields (MSY) or determine allowable catches for most commercial fish species using traditional methods. The development of data-limited methodologies that can evaluate the status of fish stock, biological reference points and biomass in using a small amount of available data and biological information, has become the focus of global regional fishery management organizations and fishery resource scholars. Based on bibliometric analysis, we summarize and review the model structure, model input and output, assumptions, sources of uncertainty, and analysis methods of the following catch-based and length-based assessment models: catch-maximum sustainable yield (Catch-MSY), depletion based stock reduction analysis (DB-SRA), depletion corrected average catch (DCAC), simple stock synthesis (SSS), an extension of Catch-MSY (CMSY), length-based spawning potential ratio (LBSPR), length-based integrated mixed effects (LIME), length-based Bayesian (LBB), and length-based risk analysis (LBRA). The focus and direction of future research are proposed based on this analysis. The results indicate that owing to the characteristics of data-limited fisheries, stock assessment is still in the development stage. If the catch and body length data are available, a model that integrates the two types of data, such as the LIME model and the SSS model, should be considered. It is suggested that the following work should be carried out in future research: (1) to actively carry out long-term, multi-sea, and full-coverage independent surveys to obtain representative sample data; (2) further optimize the existing data-limited methods and comprehensively consider the effect of various factors on the evaluation results; (3) complete and accurate basic biological research is conducted to obtain more accurate historical biological information, thereby reducing the uncertainty of the evaluation results; (4) carry out simulation testing research based on the catch-based model and length-based model to improve the tolerance of the model to statistical bias and data quality issues.