Splitting a PDF may result in a larger total file size because fonts
and other resources may be replicated in each file. Still this should
not result in a big increase in file size, so the most likely problem
is in your PDF processing code. How are you 'marking up specific pages
and resaving the file'? Do you save the resulting file with
SDFDoc.SaveOptions.e_remove_unused or e_linearized flag?
'pdftron.PDF.Optimizer' can be of use before the split operation
because it could eliminate duplicate images (e.g. the logo), fonts,
and other resources. You could also use the optimizer to decrease the
image resolution (down-sample/resample) to change image compression
etc. For a sample code showing how to use the PDF optimizer please see
http://www.pdftron.com/pdfnet/samplecode.html#Optimizer.