"I feel like a for loop might be what I need, but I'm not sure how to apply that here"
Yes, a loop is a good idea. But unfortunately those variables names are getting in the way:
As soon as you have lots and lots and lots of separate variables in the workspace with meta-data in their variable names, then you have painted yourself into a really ugly corner. A corner that you should not be in. Because then you have forced yourself into writing code that either requires lots of copy-and-pasting (as you show above) or into writing code that requires evil string evaulation. As you point out, neither of these is efficient. Best avoided.
With your example variable names, 06_05, 06_04, 06_05 are meta-data. Do NOT have meta-data in variable names. Use one array with indexing. Strore any meta-data (which is data in its own right) in variables, not in variable names.
"how do I create a list of tables (make a matrix) that the function can reference, instead of rewriting the function again and again."
So far you have not told us the most important information: how did you get all of those structures into the workspace? That is the correct place to fix this data design. For example, if they are LOADed from MAT files then always LOAD into an output variable (which is a scalar structure):
S = load(..);
and then access its fields (e.g. using FIELDNAMES and dynamic fieldnames). Then you can merge all of your data into one array and use indexing. Using loops with indexing is simple and very efficient.
"It defeats the purpose of trying to make everything more efficient"
The data design (with meta-data in the variable names) is the cause, it forces you into writing fragile and less efficient code. If you are lucky you might be able to import that data in a better way (e.g. LOADing into an output variable).
The best solution would be to design that data without any meta-data in the variable names.
