if you know the surfactant molecular size and you can determine the size of the aggregates formed, you could estimate the number of surfactant molecules in the aggregates. Afterthat, you could relate the fluorescence results to the calculations.
What kind of surfactant, at which concentration and what is the environment. Could there be an analytical error? Any indication of aggregation by a second method?
In a recent work, we could determinereliable aggregation numbers of some pluronic block copolymer (EOx-POy-EOx) using differential scanning calorimetry. However, some conditions have to be met in order to apply this method. Have a look at: http://pubs.acs.org/doi/abs/10.1021/la303599d
Many good techniques, but all have suitability for certain concentration ranges and contrast between free and micellised surfactant. In my lab we would use small angle scattering or PFG NMR.