I've read your question and have been mulling it over for awhile. Maybe some other clever person can figure a way for you build the find request you want manually, but I can't think of a way to do this.
I can think of a way to use a script to eliminate duplicate entries once you have generated the found set that displays all the required records (except it still has duplicate entries).
In the script:
1) Sort your found set so that records with matching values are adjacent. You also will want to sort it so that where there are duplicate entries, the instance you wish to retain is listed first (or last) in each cluster of matching records.
2) loop through your found set and use omit to remove duplicates from the found set (This part can be tricky as you will need to write the script to correctly handle what happens if the last record needs to be omitted. Omit moves you to the next record unless you are omitting the last one.)
You can use a variable or a global field to store the value from the previous field so you can compare it to the current record to determine whether it should be omitted.
Once you have the script working for you, you can either put a button on the layout that omits duplicates or you can include it as part of a larger script that takes user input and performs the find for them.
Hope that helps.