You are right, the radius in Cloudy and pyCloudy is the inner radius of the nebula, that is the distance from the ionizing source. It is in log of cm in pyCloudy, for example 16 (that correposnds to 1e16 cm).
The young O and B stars responsible for ionizing UC H
II
regions are so deeply embedded in molecular cloud cores that all of
their UV radiation is absorbed by dust and reradiated in the IR.
Churchwell, Wolfire, & Wood (1990, hereafter
CWW) and
Wolfire & Churchwell (1994) have modeled the 1 mm to 1

m flux distribution of G5.89-0.39 by using the spherical radiative transfer code of
Wolfire & Cassinelli (1986). The
Mathis, Rumpl, & Nordsieck (1977) grain-size distribution was used with 25 discrete grain sizes in the range 0.005 to 0.25

m
for both graphite and silicate grains. For G5.89-0.39, they found that a
dust shell of outer radius of order 1 pc, inner radius approximately
0.03 pc, and constant density of 1.7
× 10
-19 gm cm
-3
produced a satisfactory fit to the observed flux distribution. A large
dust-evacuated cavity of radius 0.03 pc was required to be consistent
with the observed low-IR fluxes. Average dust temperatures ranged from
25 K at the outer radius to about 290 K near the inner radius, with most
of the decrease in temperature occurring between 9
× 10
16 and 2
× 10
17 cm.