A flat table contains information that is duplicated many times.
Like selling a car at a dealership the bussiness would keep a list of every vehicle sold and who sold it. The salesman name and his phone number would appear many times in the table(once for every car he sold) . The dealer would also want info on the saleman like home address and tax ID number ect. There would be no need to have this information in an inventory table.
Relational databases allow sharing of information between tables by referencing a unique key from one table in another
Access can organize groups of related data in tables. For example you can have a list of car salesmen at an auto dealer listed in an employee table. This table would contain allsorts of related data to each employee like home address, full name TaxID number ect...
The same dealrship would also have a detailed list of cars in their inventory. Again with related data such as make, model, year, new or used ect.
The power access has is that it is a relational database system. You can link the above tow tables together and relate specific information like what salesman sold which car by including the salesmans employeeID number in a field in the inventory database.
This means that employee information is entered once in the employee table but may be used many times in the inventory table simply by using a single employee number as to who sold the car.
A query may then be run listing car information and then linking in to the employee table and getting specific information like name and work phone number from teh employee table.
If teh employee gets a new phone number you would just have to change a single record in the employee table rather than having to change information in every single inventory record for every car he sold.