Table Scans - More Rows Brought Back Than In Table

I have a query which is quite complex. It's based on a set of data in
a complex view which takes the data from several tables.

In this complex query, if I allow the various parts of the query to
work on the results of the view (MISView), it can take 15 minutes to
run (eek !), however, if I create a temporary table with the data from
the view and then use that for the remainder of the query, it runs in
approx 20 seconds.

Now, I have examined the execution plan (my new favourite toy) and
there is a difference (as expected). However when looking at the part
of the query that takes up most of the time, it shows that it brings
back 109,645,866 records from a table (Credit) that contains 13,002
records. This table is one that is referenced in the view (MISView)
which contains 13,653 records and does get some of it's data from the
table which is scanned (Credit).

For the record, we don't have any tables with over 100,000 records in,
so 109 million rows is going some for us. The part of the query that
runs slow does reference another copy of itself but this is necessary
for the equation that is being run.

Now I'm OK with why it's doing the table scan, but why does it bring
back substantially more data than is in the table ? Is it some
multiple of the number of records that it's trying to work out. I
assume it tries to run a seperate plan for the view as part of it's
process.

Ideally, I'm still going to go down the route of the temporary table,
but I would like to understand more about what it does first as I
don't like leaving things unanswered.

Any help would be appreciated.