First, a Compound key is tow or more FIELDS in a table that are used to make a Primary Key. A unique index is used to enforce uniqueness on a combination of fields without making those fields a PK. So lets give an example. You have a order system. So you have a order details table that records the items ordered in that order. The ProductID would not be unique in that table since different orders may include the same product. So the combination of OrderID and ProductID could be unique. So to prevent the data entry person from entering multiple line items for the same product you would create a unique index on the combination of the 2 fields. However you might still want to use a surrogate key to make it easier create joins with that table.
There is always some combination of fields that make a record unique, but using that combination as a primary key can be very cumbersome to create relationships.
Second, a surrogate key is a system generated key like an autonumber. Using a fieldname of suroogatekey makes no sense and was probably done just for illustration.
Third, I'd have to see the context about the quote; 'Surrogate keys avoid choosing a natural key that might be incorrect.'. But it is possible to select a compound key that is not unique. Going back to my example, lets say Products are further identified by size. So you would need the size to make it unique.
Fourth, every table should have a primary key. How else are you able to identify which record is being referred to in a related table. That is the purpose of keys. Access identifies the primary key in a table by using the key symbol.
Finally, the problem with a surrogate key is that, while it ensures uniqueness in a record it doesn't prevent duplication. Lets say you have a customer table. If you use a surrogate key the data entry person could enter a customer multiple times unless there is some check to make sure that doesn't happen.