Hello Karl,
1) I suggest that you rescale your problem to get rid of these nasty powers of ten. You can plot your problem from 2.5 to 2.8 rembering to apply a factor of 10^(-7) later on.
2) What do you mean by gaussian function? A normal distribution has an area of 1, so if you make it narrow it has to become high, meaning it is becoming much grater than 1.
3) If you want a general Exp[ - x^2 ] - function you should provide directly for amplitude, width and position.
I changed your code a bit which could give you a hint how to achieve what I think you want to do.
Manipulate[
Plot[
{
aa Exp[-bb (\[Lambda] - x0)^2],
1/(1 + (\[Pi]*\[Sqrt]R)/(1 - R)*
Sin[(2 \[Pi]/\[Lambda])*L]^2)}, {\[Lambda], 2.60*^-7,
2.70*^-7}, PlotRange -> {0, 1},
AxesLabel -> {"Wavelength [nm]", "Transmission"}],
{R, 0.85, 0.99},
{L, 0.00001, 0.001},
{{aa, .5}, 0, 1},
{{x0, 2.65 10^(-7)}, 2.6 10^(-7), 2.7 10^(-7)},
{bb, 10^(18), 10^(19)}]