Question: Fitting a Curve (nonlinear fitting)

Ok guys this is the problem i have a matrix wich is really the result of a semivariogram so it is like this 
thi is my semivariogram so i have to find the best model and fit it to it, i can't just throw a polynomial it has to be
spherical= h:= C((3/2)*x/a-(1/2)*x^3/a^3) 
expotencial=
Lineal= y:= p*x
or y:=p*x^a

so i want the maple to show me the variogram plot and the i select wich one of the models i will apply.
the problem is that i don't know how to apply the models so started something that could maybe lead to the values of C and A for the spherical and the xpotencial becouse the others too are easy.
------------------------------------------------------------------.--------------------------------------------------------
restart;
with(plots);
with(CurveFitting);
m3 := Matrix([[0, 0], [1, 2], [2, 5], [3, 7], [4, 10], [5, 14], [6, 15], [7, 15.4], [8, 15.8], [9, 16], [10, 16.4]]);
plot(m3);
Upsilon := C(1-exp(-x/a));
Upsilon := C((3/2)*x/a-(1/2)*x^3/a^3);
for C to 16 do
for a to 100 do
for x to 11 do
H := m3(x, 2);
R[C, a, x] := abs(Upsilon-H)
end do
end do
end do;
for r to 16 do
for j to 100 do
N[r, j] := sum(R[r, j, X], X = 1 .. 11);
if N[r, j] < 55 then
print(r, j)
end if
end do
end do;
k := 14*(3/2*((1/16)*xx)-(1/2)*xx^3/16^3);
multiple(plot, [m3, color = "Green"], [k, xx = 0 .. 11, color = "Blue"])
----------------------------------------------------------...................---------------------------------------
so in this code i rest the values of the model and the values of the matrix for differents C and a so i pick the min of the Sum to see wich C and a give me the min value but i didn't work quite well first is difficult to get the C and a and sometimes the new curve is just not similir to the semivariogram (Matrix) please help i tried 
k := NonlinearFit(C((3/2)*x/a-(1/2)*x^3/a^3), X, Y, x)
and 
k:=LeastSquares`(m3, v, curve = C*(1-exp(-x/a)))
but they don't let me because they aint linear on the parameters
please help 
if you are interested in this topic here is what i am trying to do but explained really well is for geological matters
http://www.kriging.com/PG1979/index.htm#Chapter_2+Part1

if want to help me use this matrix 
m3:= Matrix([[0, 0], [1, 1], [2, 3], [3, 5], [4, 7], [5, 8], [6, 9], [7, 10], [8, 10.5], [9, 10], [10, 9.8], [11, 9.8], [12, 10.5], [13, 10.89], [14, 11.2], [15, 10.6], [16, 9.85], [17, 9.45], [18, 9.77], [19, 10.15], [20, 10.7], [21, 10.04], [22, 10], [23, 10.85], [24, 10.7], [25, 10.94], [26, 10.81], [27, 10.33]])
for that matrix the spherical model is the best i know

goodbye