Understanding nutrient quantity and quality in forest soils is important for sustainable management of forest resources and maintaining forest ecosystem services. In this study, six soil nutrient indicators, including soil organic matter (SOM), total nitrogen (TN), total phosphorus (TP), available nitrogen (AN), available phosphorus (AP), and available potassium (AK) were measured in five different aged stands of Chinese fir forests in subtropical China. A structural equation model (SEM) was developed based on these soil nutrients indicators in order to better evaluate the soil fertility index (SFI) in these studied forests. The results show that soil nutrient contents changed with the soil depth in different age groups. The SOM decreased in a specific order: over mature > mature > near mature > middle > young stands. The TN content of the soil gradually decreased with increased soil depth throughout all age groups. The SEM indicated that the TN had the highest weight of 0.4154, while the TP had the lowest weight at 0.1991 for estimating the SFI. The weights of other indicators (AN, SOM, AP, and AK) ranged 0.2138–0.3855 in our study. The established SEM satisfied the fitness reference values and was able to accurately describe the forest soil nutrient status through the SFI. The overall SFI values were significantly higher in over mature stands than in young-aged stands and in topsoil than in deeper soil in all examined forests. Soil TN, AP, and AK were the most important nutrient indicators to the evaluation of the SFI in the study sites. The results confirmed that the SEM was suitable to estimate the weights of the SFI and better describe the soil nutrient status in forests. Our research provides an innovative approach to assess a soil nutrient status and soil fertility and provides a scientific basis for accurate implementation of soil nutrient assessment in forest ecosystems.