Knowing that the accuracy of a spatial datum depends on both the precision of the position and the thematic precision. What approach do you use for assessing accuracy for classification and modeling?
I would create a testing dataset based on visual interpretation of satellite images, use can use ROI option in ENVI. Then use it for testing accuracy instead of the ROI used for classification.
Try to select well distributed and unbaised ROI and independent of locations used in training the classfication.
For modeling:
This answer may be very subjective to the modelling approach.
1) if your model produce probabilities or soft results, go with ROC curve analysis.