I was able to find a way to do this (open and georeference) with separate files for VIIRS. Consider the example code below for two H5 files:
-----------------------------------------
; open image data
file=’SVI01_npp_d20170807_t1931133_e1943460_b00001_c20170807204544199000_ipop_dev.h5’
raster_rad=e.openraster(file, DATASET_NAME='/All_Data/VIIRS-I1-SDR_All/Radiance')
raster_rad=e.openraster(file, DATASET_NAME='/All_Data/VIIRS-I1-SDR_All/Reflectance')

;Open geometry raster that holds the x/y information:
Loc = ‘GIMGO_npp_d20170807_t1931133_e1943460_b00001_c20170807204544199000_ipop_dev.h5'
lat = e.OpenRaster(Loc, DATASET_NAME='/All_Data/VIIRS-IMG-GEO_All/Latitude')
lon = e.OpenRaster(Loc, DATASET_NAME='/All_Data/VIIRS-IMG-GEO_All/Longitude')

; Get the reprojection task from the catalog of ENVITasks
Task = ENVITask('ReprojectGLT')

; Define inputs
Task.Input_Raster = raster_rad
Task.Latitude_Raster = lat
Task.Longitude_Raster = lon

; Define output raster URI
Task.Output_Raster_URI = ‘radiance_geog.dat'

; Run the task
Task.Execute

-------------------------------------------

I also found in my notes that sometimes the lat/lon bands have 'background' values of -999.3. In that case, it can cause issues with reprojection, so I would open them with the data ignore value (div):

lat = e.OpenRaster(Loc, DATASET_NAME='/All_Data/VIIRS-IMG-GEO_All/Latitude', DATA_IGNORE_VALUE=-999.3d)
lon = e.OpenRaster(Loc, DATASET_NAME='/All_Data/VIIRS-IMG-GEO_All/Longitude', DATA_IGNORE_VALUE=-999.3d)

If you want to open more than just the data bands, you can also use the 'Template' open to OpenRaster. When you open your first data set and select all the bands of interest, you can use the 'Save Template' in the dialog. Then when opening similar files, use OpenRaster with template=myDataTemplate.

I was able to use the ReprojectGLT task to successfully reproject MODIS L1b data. I still receive an arithmetic error: Floating divide by 0, but it does produce the reprojected raster dataset. However, I would like to be able to control the output pixel resolution. Since MODIS resolution is significantly lower near the edge of swath compared to the center, the median value that is automatically selected throws away a lot of useful information near the center of the scene. I may have to resort to the classic routines after all.