The depth-specific zinc (Zn) and copper (Cu) maps with high resolution (i.e., ≤10 m) are important for soil and forest management and conservation. The objective of this study was to assess the effects of easily accessible model inputs, i.e., existing coarse-resolution parent material, pH, and soil texture maps with 1:1 800 000–2 800 000 scale and nine digital elevation model (DEM)-generated terrain attributes with 10 m resolution, on modelling Zn and Cu distributions of forest soil over a large area (e.g., thousands of km²). A total of 511 artificial neural network (ANN) models for each depth (20 cm increments to 100 cm) were built and evaluated by a 10-fold cross-validation with 385 soil profiles from the Yunfu forest, South China, about 4915 km² areas. The results indicated that the optimal models for five depths engaged five to seven DEM-generated attributes together with three coarse-resolution soil attributes as inputs, respectively, and accuracies for estimating Zn and Cu varied with R² of 0.76–0.85 and relative overall accuracy ±10% of 74%–86%. The produced maps showed that DEM-generated sediment delivery ratio, topographic position index (TPI), and aspect were the most important attributes for predicting Cu, but flow length, TPI, and slope were for Zn, which heavily affected Zn and Cu distributions in detail. Boundaries of three coarse-resolution maps were still visible in the generated maps indicated that the maps affected the distributions of Zn and Cu in large scales. Thus, the modelling method, i.e., developing ANN models with k-fold cross-validation, can be used to map high-resolution Zn and Cu over a large area.